Patient selection

A retrospective chart review was performed for all patients who had implants placed in the atrophic posterior mandible between January 1, 2009, and February 1, 2012, in one private dental practice. Five years after implant placement, patients were contacted and invited for a comprehensive clinical and radiographic examination (including periapical radiographs of the implants and their restorations). Each patient gave informed consent to participate in the study. The study protocol was approved by the Ethics Committee of Area Vasta Romagna and IRST, and was conducted according to principles stated in the Helsinki Declaration.

The following patient inclusion criteria were applied: (1) vertical bone atrophy in the posterior edentulous mandible (a residual bone height between 7 and 9 mm above the inferior alveolar nerve measured on a preoperative cone beam computed tomography (CBCT)); (2) treatment involving the placement of at least two short implants (6 mm) in the native bone or standard-length implants (≥9 mm) in ridges augmented with mandibular block bone grafts supporting a fixed partial denture; (3) delayed implant loading; (4) patients enrolled in a regular maintenance care programme; (5) a minimum follow-up period of 5 years from implant placement; (6) availability of a baseline peri-apical radiograph taken on the day of implant surgery.

The patients were excluded if were diagnosed with systemic conditions known to alter bone metabolism, such as cancer requiring chemotherapy or facial radiotherapy, intravenous amino-bisphosphonates for metastatic bone diseases, uncontrolled diabetes mellitus, immunosuppression or immunodepression and smoking habit of more than 20 cigarettes per day (all pipe or cigar smokers were also excluded).

Surgical protocol

All surgeries were performed by a single surgeon under local anaesthesia only or local anaesthesia with oral sedation (triazolam, 0.125 mg or 0.25 mg). All patients received prophylactic antibiotic therapy: 1 g of amoxicillin–clavulanic acid (or clindamycin 500 mg if allergic to penicillin) starting the night before the intervention, and twice per day for a total of 7 days, and pain medication (ibuprofen, 600 mg as needed every 6–8 hours). They were also asked to rinse with 0.2% chlorhexidine digluconate (every 12 hours starting 1 day preoperatively for 2 weeks).

Augmentation group

A midcrestal incision on the edentulous ridge was made. The incision was extended via the gingival sulcus of the adjacent tooth. Mesial and distal releasing incisions were performed, and full-thickness flaps were reflected to completely expose the atrophic ridge and identify the mental foramen. The exposed alveolar bone was cleaned of all soft tissue and several decortications holes were made using a small round bur. Bone harvesting was performed from the same side of the ramus with a piezoelectric surgical device (Piezosurgery, Mectron, Carasco, Italy). The block grafts were shaped according to the morphology of the defects and rigidly fixated to the residual bone with titanium miniscrews (Cizeta Surgicals, Bologna, Italy). Anorganic bovine bone particles (Bio-Oss, Geistlich Pharma AG, Wolhusen, Switzerland) were added at the periphery and over the bone grafts. The reconstructed sites were then covered with a resorbable collagen membrane (Bio-Gide, Geistlich Pharmaceuticals). After careful periosteal releasing incisions in order to obtain absolute tension-free closure, the flaps were sutured in two layers with a combination of horizontal mattress and interrupted sutures. The sutures were removed after 3 weeks, and the patients were then checked at 2 months after surgery. Provisional removable prostheses were not allowed during the healing period. Four to five months after augmentation, a CBCT scan was retaken to plan implant placement. The flap outline was similar to the first surgery without vestibular releasing incisions. Following mucoperiosteal flap elevation and debridement, the bone block fixation screws were removed, and standard-length implants (Astra Tech OsseoSpeed, DENTSPLY Implants, Mannheim, Germany) were placed with the top of the platform flush with the facial alveolar crest ( Fig. 1 ).

Posterior mandibular atrophic site treated with a bone block augmentation and delayed standard-length implant placement. (A) Preoperative panoramic radiograph. (B) Preoperative cone-beam computed tomography scans. (C) Preoperative occlusal view of the atrophic left posterior mandible. (D) Fixation of the bone block graft harvested from the ramus with an osteosynthesis screw. (E) Particulate anorganic bovine bone graft around block graft. (F) Collagen membrane over the block graft. (G) Panoramic radiograph at 4 months from bone graft procedure. (H) Implant placement at mandibular left first and second molar positions. (I) Periapical radiograph at implant placement. (L) Five-year periapical radiograph demonstrates some crestal bone remodelling around both implants.

Short implant group

After a crestal incision and flap reflection, implant osteotomy sites were prepared according to a standard drilling sequence. To minimize the risk of inferior alveolar nerve injury, the drills were used with adjustable stop devices that were always set at least 1 mm shorter than the radiographic working length above the canal. Two to three short implants (6 mm) (OsseoSpeed, DENTSPLY Implants) were placed in each patient ( Fig. 2 ).

Posterior mandibular atrophic site treated with short dental implants. (A) Preoperative panoramic radiograph. (B) Preoperative cone beam computed tomography scans. (C) Occlusal view of the left posterior atrophic mandible. (D) Two 4 × 6 mm implants were placed in the second premolar and first molar positions. (E) Postoperative periapical radiograph. (F) Five-year periapical radiograph demonstrates stable bone levels for both implants.

Prosthetic protocol

All implants in both groups were left submerged for 2–3 months. Definitive titanium–composite, or zirconia–ceramic FPDs were delivered 8–10 weeks after healing abutment connection; they were either screw-retained or cemented on customized computer-aided design/computer-aided manufacturing titanium abutments. All patients were kept under regular maintenance care and received full-mouth scaling every 6 months.

Outcome measures

The 5-year follow-up examination was conducted by two investigators (C.F. and E.C.), and included an update of the medical and dental history, a prosthodontic examination, and a peri-apical radiograph of all implants that fulfilled the inclusion criteria.

The following outcome measures were assessed.

• •

  Implant failure: defined as presence of any mobility of the individual implant and/or any situation dictating implant removal.

• •

  FPD failure: defined as planned FPD that could not be placed as planned due to implant failure(s), loss of the FPD secondary to implant failure(s), and replacement of the FPD for any reasons.

• •

  Surgical complications: including haemorrhaging during and after surgery, soft-tissue dehiscence, infections, fistulae, sensory loss, and partial or total bone graft resorption.

• •

  Biological complications: including peri-implant mucositis, and peri-implantitis. Peri-implant mucositis was defined as a heavily inflamed soft tissue without bone loss, and peri-implantitis was defined as a bone loss of more than 3 mm with suppuration, heavily inflamed tissues, or fistulas .

• •

  Prosthetic complications: including FPD detachment, screw loosening, and fracture of the screw, framework, or occlusal material.

Peri-implant marginal bone levels (MBLs) were evaluated on digital periapical radiographs (Digora Optime; Soredex/Orion Corporation, Helsinki, Finland) taken at implant placement, implant loading, and 5 years after implant placement. For standardization, a film holder-beam aiming device (Dentsply Rinn, Elgin, IL, USA) was applied according to the long cone paralleling technique. The radiographs were taken with the film placed parallel to the implant axis and the radiographic beam directed perpendicular to the implant . Customized positioners could not be used because of the retrospective character of the study. An image-analysis software (Digora for Windows, ver. 2.1; Soredex/Orion Corporation) was used to measure the linear distance between the implant platform and the most coronal level of the bone deemed to be in contact with the implant surface with an on‐screen cursor at 5× magnification. In cases where the implant platform was below the margin of the crestal bone, that is subcrestal, the value was considered as zero. To make calibrated measurements, an object of known size, the known diameter of the implant, was used. To adjust the measurements for any magnification error, the following equation was used to determine the correct MBLs: measured MBL × (actual implant diameter/measured implant diameter) . Relative mesial and distal bone height measurements were made to the nearest 0.01 mm and were averaged at a patient level. The change in MBL was calculated by subtracting the follow-up level from the baseline level at implant placement. Furthermore, at FPD placement, the crown/implant ratio was assessed for each implant, dividing the length of the restoration and the length of the implant. Radiographic measurements were performed by one experienced examiner (G.C.), who was calibrated and subjected to an intra-rater agreement test.

Statistical analysis

All data analyses were performed using a statistical software package (GraphPad InStat, GraphPad Software Inc., San Diego, CA, USA). Descriptive statistics were expressed as a frequency and percentage, and means and standard deviations, as appropriate. The patient was the statistical unit of the analyses. Differences in the proportion of patients with failures and complications were compared between the groups using Fisher’s exact test. The Friedman test was used to detect intra-group differences, and Mann-Whitney test was used to detect between-group differences in MBL values. The level of significance was set at 0.05 for all comparisons.