Dental implants with versus without peri-implant bone defects treated with guided bone regeneration

Background The guided bone regeneration (GBR) technique is highly successful for the treatment of peri-implant bone defects. The aim was to determine whether or not implants associated with GBR due to peri-implant defects show the same survival and success rates as implants placed in native bone without defects. Material and Methods Patients with a minimum of two submerged dental implants: one suffering a dehiscence or fenestration defect during placement and undergoing simultaneous guided bone regeneration (test group), versus the other entirely surrounded by bone (control group) were treated and monitored annually for three years. Complications with the healing procedure, implant survival, implant success and peri-implant marginal bone loss were assessed. Statistical analysis was performed with non-parametric tests setting an alpha value of 0.05. Results Seventy-two patients and 326 implants were included (142 test, 184 control). One hundred and twenty-five dehiscences (average height 1.92±1.11) and 18 fenestrations (average height 3.34±2.16) were treated. At 3 years post-loading, implant survival rates were 95.7% (test) and 97.3% (control) and implant success rates were 93.6% and 96.2%, respectively. Mean marginal bone loss was 0.54 (SD 0.26 mm) for the test group and 0.43 (SD 0.22 mm) for the control group. No statistically significant differences between both groups were found. Conclusions Within the limits of this study, implants with peri-implant defects treated with guided bone regeneration exhibited similar survival and success rates and peri-implant marginal bone loss to implants without those defects. Large-scale randomized controlled studies with longer follow-ups involving the assessment of esthetic parameters and hard and soft peri-implant tissue stability are needed. Key words:Guided bone regeneration, peri-implant defects, dental implants, marginal bone level, success rate, survival rate.


Introduction
Horizontal alveolar bone defects often result in a dehiscence or a fenestration defect exposing part of the implant surface (1,2). Several clinical studies have shown that at least 1mm of bone width buccal and lingual to the implant surface is needed to ensure long-term bone coverage and therefore implant success (3,4). When this is lacking at the moment of implant placement, guided bone regeneration has been proposed to augment the bone width in a single simultaneous surgical intervention (5,6). An important issue is whether or not implants placed in sites associated with bone regeneration provide survival and success rates similar to those of implants placed in sites with sufficient native bone (7). Although there are a variety of case series studies on bone regeneration, three systematic reviews (1,8,9) on implant survival in sites regenerated with GBR identified only three studies (10)(11)(12) that compared implants with peri-implant defects that required bone grafts versus implants entirely surrounded by pristine bone as control implants in their analysis. The purpose of the present study was to retrospectively evaluate whether or not implants associated with bone regeneration due to peri-implant defects show the same survival and success rates as implants placed in native bone without such defects, and to evaluate long-term outcomes of implants with dehiscences and fenestrations treated with guided bone regeneration with a minimum follow-up of three years post-loading.

Subject and site inclusion criteria:
A minimum of two dental implants, one with a dehiscence or fenestration bony defect during implant place-ment treated with particulate bone graft and resorbable membranes (Test Group), and another implant entirely surrounded by bone (Control Group

Material and Methods
-Patient selection A retrospective controlled clinical study was made of patients with a minimum of two dental implants, one implant demonstrating a dehiscence or fenestration bony defect with exposed implant surface during implant placement and so undergoing simultaneous particulate bone grafting with resorbable membranes (test group), and the other implant site entirely surrounded by bone (control group). All implants had to be left submerged. Patients were treated between January 2005 and December 2009 at the Oral Surgery Unit of the University of Valencia in Spain and were monitored annually for a minimum of 3-years post-loading. Patients were given full information about the surgical procedures and duly signed informed consent forms. Preoperative analysis included registering complete medical histories and performing clinical and radiographic examinations. Inclusion and exclusion criteria are detailed in table 1. The present study is reported in accordance with the STRO-BE (Strengthening the Reporting of Observational Studies in Epidemiology) statement (13).
-Pre-Operative Evaluation Thorough medical histories, clinical examinations and panoramic radiographs were performed in all cases. Cone-beam computed tomographic scans were obtained to assess the availability of bone whenever the surgeon considered this necessary. Periodontal treatment was provided whenever necessary to control inflammation prior implant placement surgery. Within 10 days of the implant placement surgery, a full mouth professional prophylaxis appointment was scheduled.
-Surgical procedures Surgical procedures were performed by the same surgeon with an extensive experience in regenerative procedures. All procedures were performed under local anesthesia using 4% articaine 1:100,000 adrenalin (Inibsa, Lliça de Vall, Spain) and intravenous conscious sedation with 1% propofol solution, administered by an anesthesiologist. All patients received antibiotic prophylaxis (amoxicillin and clavulanic acid, 1g every 8 hours starting 1 day preoperatively (14)). An initial incision was made slightly palatal/lingual of the alveolar crest. One or two releasing incisions were made and a mucoperiosteal flap was raised. The exposed alveolar bone was curetted to remove all soft tissues. To enhance primary stability, drills and osteotomes were combined to prepare implant beds. TSATM implants with Avantblast surface (Phibo Dental Solutions S.L., Sentmenat, Barcelona, Spain) were inserted using standard procedures following the manufacturer's guidelines. These implants have a polished surface portion of 1.5 mm. All implants were placed with adequate primary stability (≥35 Ncm). In implants that did not need bone regeneration, bone width from the implant head to the facial plate was over 1.5 mm. In implants that needed bone regeneration, autologous bone grafts harvested from the conformation of implant beds during drilling and was adjusted to the bone contour. When the autologous bone obtained was of insufficient quantity to cover the periimplant defects, synthetic bone (Kera-OsTM, Keramat, Coruña, Spain) was added. Grafted bone was protected with a textured collagen membrane (Lyopstic, B Braun, Aesculap, Germany). Periosteal incisions were made to allow flap mobilization and tension free primary wound closure. Implants were left submerged. Flaps were closed with horizontal sutures using Polisoft® 4/0 sutures (Sweden & Martina, Due Carrare, Italy) (Figs. 1-10).       thereafter. Patients were instructed in adequate hygiene maintenance and a soft diet was recommended for eight weeks. Patients were not allowed to use removable prostheses for three weeks after bone grafting surgeries. A soft diet was recommended for one week and patients were instructed to avoid brushing or any other trauma to the surgical sites. Sutures were removed two weeks after surgery. Second surgeries were performed two or three months after implant placement and final fixed prostheses were placed one month later.
-Data collection and follow-up All data collection was carried out by a single trained clinician, different from the surgeons or the prosthodontist, following a pre-established protocol. All patients were included in a maintenance program involving annual examinations and professional prophylaxis. Patient age (at implant placement), gender, hygiene and smoking habits (none / <10 cigarettes per day / 10-20 cigarettes per day) were registered. For each implant, the position, and the type and dimensions of defects (dehiscence/fenestration) were registered. Defects were measured using a millimetric periodontal probe (Hu-Friedy UNC, Chicago, IL, USA) placed parallel or perpendicular to the long axis of the implant. Measurements were recorded to the nearest 1mm mark. The use of temporary prostheses (yes or no), definitive prosthesis design (single, partial or complete -fixed or overdenture) and type of prosthesis (cemented or screwed) were recorded. All patients were included in a maintenance program involving annual examinations and occlusal adjustment was performed when necessary.

Results
A total of 129 patients with submerged dental implants placed with dehiscences and fenestrations and treated with particulate bone graft and implants without these bony defects in the same patient were included. Eleven patients were excluded as a result of failing to attend the scheduled recall visits and 20 because control implants were not submerged. These exposures did not exceed 3mm in diameter. In these cases, 0.2% chlorhexidine gel was prescribed three times daily over the exposed membrane for 6 weeks after surgery. All sites re-epithelialized uneventfully.  for the control group. Differences between both groups were not statistically significant (p=0.893). No significant differences were found between implant failures/bone loss and patient factors (age, gender, smoking habit) or implant variables (location, type defect, defect dimensions, prostheses). However, implant failures were more frequent among patients with a regular oral hygiene (91.6%) than good (8.4%); these results showed a moderate tendency to significance but were nevertheless non-significant (p-value 0.06).

Discussion
Peri-implant bone defects occurring during the surgical implant placement procedure are a frequent problem in daily practice, one with considerable clinical relevance. For this reason, the present study set out to evaluate the long-term outcome of implants with dehiscences and fenestrations treated with guided bone regeneration with a minimum follow-up of three years. The study analyzed complications associated with guided bone regeneration, implant survival and success rates and peri-implant marginal bone loss. Soft tissue thickness and primary flap closure are important aspects of surgery for the maintenance of wound stability during healing (16). Early exposure of the membrane, with consequential bacterial contamination of the healing tissues, hinders bone regeneration, despite careful remedy with chlorhexidine applications (17,18). Nowzari and Slots (19) showed that implant sites with submerged barrier membranes remained free of cultivable microorganisms throughout their nine-month study and experienced significantly more osseous healing than sites with prematurely exposed membranes. The application of chlorhexidine has been recommended for cases of membrane exposure (2,3). In the present study, wound dehiscence with membrane exposure occurred at twelve grafted sites in twelve patients during the early postoperative period.
There is insufficient evidence in the literature in support of the need to cover the dehiscence and fenestrations for ensuring long-term implant survival (20). To address this issue, randomized clinical trials that compare treatment with GBR vs. non-treatment of fenestrations/dehiscences should be performed. But to date, no research has been published with these precise characteristics. Dahlin et al. (21) conducted the first split-mouth study on implants (machined titanium) with fenestrations treated with or without e-PTFE membrane; there was a significant bone gain in the group treated with GBR in comparison with the non-treated group, but there was no difference in the implant survival rates between groups. The authors showed that membranes alone, without the additional aid and sustenance provided by graft materials, will allow the formation of new bone; however, it could not be demonstrated that the realization of this bone augmentation was necessary for implant survival. The present article was only able to demonstrate that dehisced implants, if properly treated with grafting materials and membranes, may lead to a survival rate of implants similar to that obtained in implants completely embedded in native bone, but it was not designed to demonstrate whether GBR is really needed in case of fenestrations/dehiscences. Only four studies have set out to determine whether the survival of implants with bone defects treated with peri-implant regenerative procedures is comparable to implants without defects in the same patient (7,(10)(11)(12). According to these studies, survival and success rates were not significantly different between implants in regenerated and non-regenerated bone. In the present study, implant survival rates were 95.7% (test) and 97.3% (control) and implant success rates were 93.6% and 96.2%, respectively, at 3 years post-loading. These results compare well with systematic reviews reporting survival rates of implants placed into sites with regenerated bone ranging from 79% to 100%, with the majority of studies indicating >90% after at least 1 year of function (9,22). The survival rates obtained in these systematic reviews were similar to those reported in the meta-analysis of prospective longitudinal studies for implant therapy with and without bone regeneration after at least 5 years (23). The assessment of bone levels around implants with buccal or lingual defects presents a methodological challenge. Schliephake et al. (24) showed that determining peri-implant bone levels from periapical radiographs, reformatted CT scans or direct magnification imaging does not provide valid data on the circumferential bone level in implants with buccal bone defects, but does generally reflect the bone level on the lingual side. Periapical radiographs were found to obtain the best correlation, though they tend to overestimate the bone anchorage of these implants. Authors such as Mayfield et al. (10) have hypothesized that distal and mesial marginal bone values reflect the buccal bone margin. Jung et al. (5) found minimal mean changes (0.03-0.13mm) of the marginal bone level at test sites and control sites between baseline, 3-and 5-year examinations. In a retrospective multi-center study, the mean radiographic bone loss over a 74-month period of implants loaded in regenerated bone was 0.64mm (0.3-0.8mm) (25). A further mean annual crestal bone loss ranging from 0.05-0.13mm after the first year of function (i.e. after abutment connection) has been reported (26)(27). In this study, mean peri-implant marginal bone loss after 3-years post-loading was 0.54 ± 0.26 mm for the test group and 0.43 ± 0.22 mm for the control group, without significant differences between groups. These results demonstrated that bone regenerated by GBR in peri-implant bone defects remains as stable over time as pristine periimplant bone. These data compare well with results reported in previous studies including implants in pristine as well as regenerated bone (10,12).
In the present study, re-entries were not performed, so it provides no evidence on the effectiveness of this technique on bone regeneration. In addition, it should be noted that all procedures were performed by the same oral surgeon with extensive clinical experience in regenerative procedures, which may limit the extrapolation of the results. Nevertheless, the lack of appropriate controls and the lack of good outcome measurements that assess aesthetic demands preclude definitive conclusions. Further long-term studies with appropriate controls and larger sample sizes with longer follow-ups should be conducted in order to confirm or reject these findings.

Conclusions
Within the limits of this study, implants with peri-implant defects treated with guided bone regeneration exhibited similar survival and success rates, and periimplant marginal bone loss to implants without these defects. Implant survival and success rates were high in both groups; marginal bone levels were within the range of published data.