The objective of this study was to evaluate the outcomes of mandibular vertical defect reconstruction with autologous bone and the use of a sub-periosteal tunnel approach in preparation for dental implant insertion. Forty-three consecutive patients with an atrophic posterior mandible were reconstructed using this method. Two thin laminae of cortical bone, obtained by splitting blocks harvested from the retromolar area, were fixed in a box-like framework containing cancellous and particulate bone. The goal was to achieve an alveolar ridge width of ≥5.5 mm and an effective bone height (EBH) of ≥10.5 mm for dental implant insertion (≥3.4 mm diameter, ≥9.5 mm length). Fifty reconstruction procedures were performed. The mean EBH was 7.1 ± 1.3 mm pre-treatment and 12.3 ± 1.1 mm post-treatment (mean increase 5.2 ± 1.4 mm). Complete graft loss was recorded in two cases; the remaining complications were minor. After a mean consolidation period of 3.5 months, 96 dental implants were placed. No failure of osseointegration was observed at follow-up (mean 32.9 months). The average bone height reduction was 0.9 mm (graft vertical resorption 17.4%). Reconstruction of posterior mandibular vertical defects using two autogenous cortical bone blocks with particulate bone between them, combined with a tunnelling technique, provided good healing with no wound dehiscence and minimum resorption of the grafted bone, favouring a substantial vertical bone gain.
Many techniques have been developed for the reconstruction of posterior mandible bone defects, to achieve a sufficient bone volume for the ideal placement of dental implants. Most authors agree on the superiority of autologous bone for mandibular alveolar reconstruction and on the need to preserve adequate soft tissue coverage in bone augmentation surgery. The most widely applied techniques include onlay grafting of block bone, generally harvested from extraoral (iliac crest, calvarial bone) or intraoral (symphysis, mandibular ramus) areas, guided bone regeneration using membranes or titanium meshes, and the sandwich osteotomy with interposed bone block graft. Although onlay bone grafts are generally recognized as the gold standard for alveolar reconstruction, intraoral bone harvesting has been associated with unsatisfactory bone quantity and unpredictable resorption of the grafts over the long term. In addition, the conventional crestal incision used to gain access to the atrophic crest to graft the onlay bone is associated with a high risk of dehiscence during the healing process, with consequent exposure, contamination, and partial or total loss of the bone graft or bone graft substitutes, thereby compromising the final result of the reconstruction.
Structurally, the ideal graft should have the thinnest possible outer cortical layer and a predominant inner cancellous layer to promote its rapid vascularization and nutrition and to strengthen its mechanical stability at the same time. Appropriate graft immobilization is also important to avoid micromotion and the consequent rupture of vascular buds, which can lead to a failure of graft incorporation into the receptor bed. The main aim in mandibular alveolar reconstruction is to graft a sufficient quantity of structurally resistant bone block to allow three-dimensional (3D) reconstruction of the defect and the ready post-fixation incorporation of the graft into the receptor bed, using a flap to cover the bone without tension of the supra-adjacent soft tissues.
To minimize the risk of soft tissue dehiscence, some modifications have been made to avoid the crestal incision, such as the utilization of a sub-periosteal tunnel access (STA) with two vertical vestibular incisions. Khoury et al. suggested that a tunnel technique would be a safer approach to ensure the integrity of the soft tissues over the graft. They also reported an original 3D grafting technique developed using autologous bone grafts harvested from the mandibular retromolar area to obtain a substantial vertical bone gain in preparation for implant insertion.
The purpose of this prospective study was to assess the clinical outcomes in terms of bone gain, success rate on long-term follow-up, and complications of a 3D reconstruction technique for the repair of vertical defects of the posterior mandible using a tunnel technique approach, with two thin blocks of cortical bone fixated at the receptor site in a box-like framework containing cancellous and particulate bone, in preparation for the placement of dental implants. The two thin blocks of cortical bone were obtained from splitting a block of autogenous graft harvested from the retromolar area.
Materials and methods
This prospective observational study included 43 patients treated for 50 vertical defects of the atrophic posterior mandible resulting from long-standing partial edentulism. These patients were treated consecutively by the authors with an autologous bone reconstruction technique and STA, in the department of oral and maxillofacial surgery of the university hospital in Madrid, Spain, between January 2010 and December 2012. The objective was to create an alveolar ridge width ≥5.5 mm and an effective bone height (EBH) ≥10.5 mm (distance from the ridge to the alveolar nerve canal), i.e., adequate bone volume for the insertion of dental implants with a diameter ≥3.4 mm and length ≥9.5 mm, thus avoiding the need for short implants ( Fig. 1 ).
Inclusion criteria were the following: a severely resorbed posterior mandible (class II–IV mandibular atrophy, according to the Cawood and Howell classification ), Kennedy class I–II partial edentulism, and patient request for implant-supported restorations. The posterior region of the mandible was defined as the area located posterior to the first bicuspid. The following exclusion criteria were applied prior to surgery: a smoking habit, poorly controlled diabetes, previous history of radiotherapy, and refusal to provide written informed consent.
Patients received antibiotic prophylaxis with 1 g/250 mg oral amoxicillin–clavulanic acid every 12 h, commencing at 1 h before surgery and continuing for 7 days. The donor and receptor sites were infiltrated with local anaesthetic (1:100,000 articaine hydrochloride with adrenalin) under either conscious intravenous sedation or general anaesthesia, as appropriate.
At the receptor site, a number 15 scalpel was used to make a single vertical incision in the vestibular gingival mucosa, mesial to the bone defect; a parallel second vertical incision was made distal to the defect. Following this, a full-thickness buccal and crestal flap was elevated to expose the reconstruction area by connecting the two incisions and creating a sub-periosteal tunnel. A corticocancellous block graft measuring ≥4 mm in width was then obtained from the ipsilateral retromolar area. For this purpose, a horizontal incision along the mandibular external oblique ridge was made from the distal gingival vertical incision, for wide exposure of the donor site ( Fig. 2 a) .
The bone graft was cut along its long axis into two thinner blocks using microsaw diamond discs (FRIOS MicroSaw; Dentsply Implants, Mannheim, Germany) ( Fig. 2 b–d). One of the graft laminae was introduced through the mucoperiosteal incision of the tunnel, placed in a crestal position supported on the anterior and posterior piers of the bone defect, and fixated with two 1.2-mm osteosynthesis screws (Stryker Leibinger, Freiburg, Germany) ( Fig. 3 a) . The length of the screws was determined prior to surgery by measuring the distance to the alveolar nerve canal on the cone beam computed tomography (CBCT) scan (taken before surgery) and taking into account the expected vertical gain. A minimum of 3 mm screw tapping into the recipient bone was needed to ensure the correct stability of the graft laminae. Next, a cancellous bone graft harvested from the retromolar site together with particulate bone obtained from the rest of the block was used to fill the space between the crestal cortical thin lamina and the native alveolar bone ( Fig. 3 b). The other thin cortical block was then placed in a vestibular position (box-like) over the particulate bone and fixed with two screws into the buccal aspect to complete the reconstruction of the defect ( Fig. 3 c). Thus, the particulate bone was packed from the vestibular side before the ‘box’ was ‘locked’ by the vestibular cortical block. The mucosal incisions were closed with a 4–0 polyglactin resorbable suture (Vicryl Rapide, Ethicon) in a single layer.
At between 4 and 5 months post-grafting surgery, the screws were removed and the dental implants were fixed under local anaesthesia. A panoramic radiograph was taken prior to implantation to determine and measure the EBH achieved and to select the implant length ( Fig. 3 d). Patients with <5 mm of keratinized gingiva at the ridge level underwent a vestibuloplasty at this time (modified Kazanjian technique); in the other cases, a crestal incision was made ( Fig. 4 ). After a 4-month osseointegration period, the implants were uncovered, gingival moulding devices were placed by crestal incision, and the prosthetic procedure was initiated.
Before bone grafting treatment, all patients underwent panoramic radiography and a CBCT, which were used to calculate the size and shape of the required graft. Between 3 and 4 months after grafting, prior to implant placement, the EBH was determined from the panoramic radiograph by measuring a vertical line from the lowest point of the residual alveolar ridge to the upper border of the alveolar nerve canal; this measurement was confirmed on CBCT images in all cases. Panoramic images were obtained using a Planmeca ProMax digital panoramic X-ray (Planmeca Co., Helsinki, Finland) with a 2.5-mm aluminium filtration and 60 kVp and 4 mA kilovoltage adjustments. The same team of technicians took all radiographs.
After completion of the prosthodontic treatment, the patients were monitored with clinical and radiological examinations ( Fig. 5 ). As all implants were placed crestally, the initial bone level was considered to be located at the height of the implant shoulder. Vertical bone resorption, mesial and distal to the implants, was calculated on a panoramic radiograph obtained at 1 year after functional implant loading. The distance from the implant shoulder to the crestal bone level was measured at each study time point. To assess bone resorption, the arithmetic mean was obtained for all measurements. The same investigator performed all measurements. The magnification was taken into account by comparing the implant length in the radiograph with the measurement recorded in the medical records. The same procedure was repeated on an annual basis.
Data were entered prospectively into a database and updated regularly. The data collected included age and gender, bone defect location, graft fixing procedure, vertical height achieved, delay to implant surgery, number of implants, implant location, length, and diameter, associated soft tissue procedures, and related complications observed at the follow-up visits. Descriptive statistics of the variables was performed using absolute frequencies and relative frequencies for qualitative variables. No analytical statistical analysis was performed because of the small number of patients enrolled in the study. The protocol was designed in accordance with the principles of the Declaration of Helsinki and the study was reviewed and approved by the local ethics committee. Written informed consent was obtained from each patient.
Fifty procedures to reconstruct posterior mandibular vertical defects were performed in 43 patients (37 females and six males). The mean age of the patients was 49.7 years (range 32–67 years). Seven patients underwent a bilateral procedure (performed simultaneously in six patients). Twenty-three of the procedures were executed on the right side and 27 on the left side. Thirty-two of the patients underwent conscious intravenous sedation and 11 underwent general anaesthesia (in cases of bilateral reconstruction or additional surgical procedures). The average operation time for a unilateral procedure was 87 min (range 74–118 min). Bone grafts were fixed with four screws in 43 procedures, while five screws were required in seven procedures (mean 4.1 screws). The follow-up period after the bone grafting procedure ranged from 22.2 to 57.4 months (mean 38 months).
The main postoperative complication was complete graft loss due to crestal lamina mobility, recorded in two cases; one was caused by an early fracture and the other by a fixation failure. Neurosensory dysfunction of the mental nerve was found at 16 operated sites at 1 week post-surgery, assessed by two-point discrimination test with sharp callipers. Recovery was complete within 1 month in 10 cases, within 3 months in three cases, and within 6 months in a final case. Minor altered sensation was still present in two patients at 1 year post-surgery. Other minor complications with no sequelae were: one infection in the donor area at 3 weeks post-surgery that responded well to antibiotic treatment, three cases of major postoperative oedema that disappeared after 2–3 weeks, three exposures of screw heads, and five small lingual exposures of the posterior angle of the graft crestal layer that were resolved successfully by drilling the exposed fragment.
The mean EBH was 7.1 ± 1.3 mm before bone grafting surgery and 12.3 ± 1.1 mm after bone grafting surgery, measured prior to implant placement. The mean amount of height gain was 5.2 ± 1.4 mm ( Table 1 ). The mean graft consolidation period before implant placement surgery was 3.5 months (range 3.3–4.4 months). A total of 96 implants were placed (XiVE model with original CELLplus surface; Dentsply Implants, Mannheim, Germany), with diameters of 3.4 mm ( n = 7), 3.8 mm ( n = 84), and 4.5 mm ( n = 5), and lengths of 9.5 mm (n = 37), 11 mm ( n = 52), and 13 mm ( n = 7). Patients received between one and three implants in the reconstructed areas (mean two implants per patient). The distribution of the fixations was as follows: first bicuspid, four implants; second bicuspid, 19 implants; first molar, 48 implants; and second molar, 25 implants. For implant placement, a crestal approach was used in nine patients and a vestibuloplasty in 39 patients. The osseointegration success rate was 100%.
|EBH pre-reconstructive surgery||7.1||7||3.5–9.2||1.3|
|EBH post-reconstructive surgery||12.3||12||10.5–14.5||1.1|
The mean follow-up after implant placement was 32.9 months (range 18.7–53.7 months). The fixture survival was 100%, and the function and stability of these implants was judged to be satisfactory. No major resorptive modification of the grafts was detected radiologically. The average graft height reduction around implants was 0.9 mm (17.4%) at the time of the study, over 1–4 years post implant functional loading (mean 31 months). The highest annual resorption rates were reported after the first year of loading (mean 0.8 mm), decreasing to 0.1 mm after the second year and to 0 mm in the third and fourth years ( Table 2 ).
|Implant site a||Number of implants||1 year||2 years||3 years||4 years|
|Mesial||Distal||Number of implants||Mesial reduction||Distal reduction||Number of implants||Mesial reduction||Distal reduction||Number of implants||Mesial reduction||Distal reduction|