This study investigated the biomechanical effects of crestal bone osteoplasty and flattening procedures carried out in edentulous knife-edge ridges to restore bone width before implant placement on the virtually placed implants using finite element methods. Three-dimensional models representing a knife-edged alveolar bone with two different crestal cortical bone thicknesses (1.6 mm, thin group; 3.2 mm, thick group) were created. Gradual crestal bone osteoplasty with 0.5 mm height intervals was simulated. Cylindrical implants with abutments and crowns were constructed and subjected to oblique loads. Maximum stress was observed at the cervical region around the implant neck. Different osteoplasty levels showed different stress values and distributions. Highest compressive stress was observed in the flat models (60.8 MPa and 98.3 MPa in thick and thin groups, respectively), lowest values were observed when osteoplasty was limited to the sharp edge (36.8 MPa and 38.9 MPa in thick and thin groups, respectively). The results suggested that eliminating the sharp configuration in knife-edge ridges improved stress and strain outcomes, but flattening the alveolar crest and/or uncovering the cancellous bone resulted in a marked increase in compressive stress and strain values in the peri-implant bone that may influence the longevity of implants placed in these ridges.
Dental implants have been widely used for prosthetic rehabilitation of partially and completely edentulous patients. Adequate available bone dimensions are considered a prerequisite for successful and predictable implant treatment . Following tooth extraction, continuous bone resorption usually takes place and results, initially, in a narrow ridge with a knife-edge form .
A relatively large percentage of narrow ridges with knife-edge configuration have been reported at the edentulous mandibular and maxillary bony crests . Treatment for completely removable dentures may require preprosthetic surgical reduction of the sharp edges to prevent painful denture pressure points at the knife-edged borders. Increasing the horizontal width of the bone prior to implant placement is required in these ridges to host the implant in the alveolar bone properly . This can be achieved by osteoplasty, performed to eliminate and flatten the thin sharp edge configuration, and/or additional bone augmentation and surgical procedures such as onlay bone grafts , guided bone regeneration horizontal distraction and sagittal osteotomies of the edentulous ridge .
When the narrow width is limited to the alveolar bone crest in ridges with adequate height, osteoplasty or flattening procedures are recommended rather than bone augmentation procedures . The local anatomy and geometry of the peri-implant bone influence the distribution and intensity of the stresses generated in the surrounding bone, because of the mechanical interlocking relationship between the implant and the surrounding bone . Excessive stresses generated around dental implants are considered to be one of the main causes of peri-implant bone loss and/or implant failure .
The effect of crestal bone osteoplasty and the flattening procedures that precede implant placement on the biomechanical behavior of dental implants has not been assessed in relation to the stress and strain generated around the dental implants that will be placed in these ridges. The purpose of this study was to evaluate the biomechanical effect of surgical reduction and flattening of the narrow alveolar bone crest on the biomechanical behavior of the future implants using three dimensional (3D) finite element analysis (FEA). Several methods have been used to evaluate the biomechanical aspect of dental implants such as FEA, strain gauges and photo-elastic models. Of these methods, FEA is considered the method of choice to calculate the stresses generated in complex geometries and to evaluate different variables simultaneously .
Materials and methods
A 3D model of an edentulous mandibular segment with knife-edge configuration was constructed. A cross-sectional CT scan for the mandibular first premolar region was selected with a bucco-lingual dimension of less than 2 mm at the alveolar crest level, which is considered as the knife-edge ridge according to P ietrokovski et al. .
The image was plotted and used to determine x and y coordinates for points that describe the outline of the external cortical bone surface. These coordinates were imported into the FEM software (ANSYS 9.0, ANSYS Inc., PA, USA) as keypoints that were connected by smooth lines/curves using a ‘spline algorithm’. The area was divided into a cancellous core surrounded by a layer of cortical bone. Two basic bone models (Knife models) were created with exactly the same external cortical bone outline, but different cortical bone thicknesses at the crestal region. The first had a uniform layer of cortical bone 1.6 mm thick (thin group), and the second had cortical bone thickness at the crestal region of 3.2 mm (thick group). The two-dimensional images were extruded in the z axis to create a 3D bone model with a mesio-distal length of 20 mm. The alveolar bone models were approximately 12.25 mm wide bucco-lingually and 28.5 mm heigh infero-superiorly.
Flattening and several levels of osteoplasty were simulated for the two basic Knife alveolar bone models by gradually subtracting the bone in the crestal region until a flattened surface was obtained in the Flat models, with 1 mm of bone on the buccal and distal sides. A vertical distance of 0.5 mm (measured from the highest point vertically) was removed each time to produce the Reduction models: R1, R2, R3, R4 and R5 ( Fig. 1 ). The reduction was performed in a curved way to preserve the bucco-lingual width and was limited to the central part of the bone segment with a mesio-distal length of 8 mm, corresponding to the standard implant diameter (4 mm) and 2 mm on both mesial and distal sides. 14 models were constructed ( Fig. 1 ). A cylindrical implant, 4 mm in diameter and 10 mm long, was placed vertically in the alveolar ridge, with a simplified abutment and crown, 6 mm in diameter and 8 mm high.
Loading and boundary conditions
All materials were assumed to be isotropic, homogeneous and linearly elastic. The interfaces between the materials were assumed bonded or osseointegrated. Young’s moduli of the cortical bone, cancellous bone, implant, and prosthetic structures were assumed to be 15 GPa, 1.5 GPa, 110 GPa and 96.6 GPa, respectively. A Poisson’s ratio of 0.3 was used for all the materials, except for the prosthetic structures, which had a ratio of 0.35 .
The models were fixed in all directions on the mesial and distal surfaces of the bone . A load of 200 N applied 30° buccal from the vertical axis was applied at the center of the occlusal surface of the crown.
All models were meshed with 20-node tetrahedral elements. A convergence test was performed to determine the number of elements to ensure the validity of calculations. A finer mesh was generated around the neck of the implant. Depending on the model, element numbers ranges were 75,175–82,793 and 90,435–98,328 in the thick and thin groups, respectively. The compressive stress and strain results were calculated and recorded for all models in the bone structures.
The maximum compressive stress and strain values in each of the 14 models are summarized in Figs 2 and 3 . On loading the implant, maximum compressive stresses were observed at the cervical region of the cortical bone in all models. In the knife-edge models, the maximum stress was noted at the mesial and distal sides of the implant neck on the sharp bone edge ( Figs 4 and 5 ). Maximum cortical bone stress values were 49.5 MPa and 66.3 MPa in the thick and thin models, respectively. A marked reduction in stress was noted when the sharp edge was eliminated in R1 models with peak compressive stress of 36.8 MPa and 38.9 MPa for thick and thin cortical bone models, respectively. The maximum compressive stress was increased progressively as the alveolar crest height became reduced. The highest compressive stress value occurred around the implant in the flat models with 60.8 MPa and 98.3 MPa in the thick and thin cortical bone models, respectively ( Figs 4 and 5 ).
In the cancellous bone, the maximum stress values were considerably lower than those in the cortical bone in all models. Peak stresses were observed at the crestal region under the cortical bone plate around the implant neck in R3, R4, R5 and Flat models of the thin group and the Flat model of the thick group. In the rest of the models, peak stresses were observed at the lingual side of the implant apex ( Fig. 6 ). The distribution of the maximum compressive strain was similar to the compressive stress in the corresponding models.