Peri-Implant Biomechanical Responses to Standard, Short-Wide, and Mini Implants Supporting Single Crowns Under Axial and Off-Axial Loading (an In Vitro Study)

Implant selection is generally based on the maximum amount of available bone. This is based on the fact that favorable load distributions exist when the greatest surface area of bone is contacted by the implant to facilitate the transfer of occlusal forces.¹  Yet, in the presence of limited alveolar bone height or diminished alveolar ridge buccolingually, the use of a standard diameter implant with adequate length and diameter is not an option.

The quantity of bone in the vertical direction and the distance between the teeth adjacent to a missing tooth are the main criteria when selecting the length and diameter of an implant.^2,3  It has been shown that the distance between an implant and natural teeth must not be less than 1.25 mm in order to leave a sufficient distance for bone and periodontal membrane for implants placed between 2 natural teeth, as a proper blood supply is necessary for osseointegration.³  Moreover, the existence of 0.5-mm-thick bone around the implants is advocated for long-term implant success.³ 

When there is insufficient bone around the implants or in the presence of severe bone atrophy, the volume of bone can be increased by bone augmentation, followed by insertion of a standard size implant.^4,5  Yet, as an alternative to bone grafting and to avoid subjecting patients to multiple surgical procedures, mini implants were suggested to be inserted in thin wiry ridges and in the presence of the narrow space between 2 natural teeth. Short implants with a wide diameter were suggested to be used in alveolar ridges with reduced height.^6–8  Moreover, in the case of bone of poor quality and quantity, some have suggested increasing the implant diameter as a way of increasing tolerance of occlusal forces, improving initial stability, and providing a favorable stress distribution to the surrounding bone.^9–11 

Development and maintenance of the implant-bone interface is dependent mainly on the control of biomechanical loads.¹²  It was stated that among the factors that affect the load transfer at the bone-implant interface are implant geometry,¹³  diameter,¹⁴  and the surface area of the implant integrated into the bone.⁸  From a bioengineering perspective, an important issue is to design the implant with a geometry that will minimize the peak bone stress caused by standard loading.¹⁵  For these reasons, the use of mini implants or short-wide implants to support single crowns has been a source of debate.^16–22 

Bone responds to alterations in loading conditions in the form of bone remodeling, according to Wolf's law, as adaptation of bone morphology is regulated by mechanical loading.²³  If the levels of stress or strain in the bone go beyond a threshold that exceeds the bone's capacity, (ie, 3000 μ€),²⁴  mechanical fatigue damage of the bone occurs, giving rise to the loss of marginal bone or even complete loss of osseointegration.^25,26 

One of the challenges in designing a dental implant system is to create a favorable biomechanical environment that prevents the surrounding bone from resorbing and/or failing under normal occlusal loads. Therefore, it is necessary to understand how the stress concentration on jaw bones is affected by different types of loading and the diameter and length of the implants.

Because of the complex geometry of the coupled bone-implant biomechanical system that prevents the use of the closed-form approach for stress evaluation, the behavior of endosteal dental implants was investigated by using numerical techniques. Recently, finite element analysis (FEA) has been widely applied to prosthetic dentistry^27,28  to predict stress and strain distributions at peri-implant regions; investigate the influences of implant and prosthesis designs,^29–32  the magnitude and direction of loads,^32,33  and bone mechanical properties^33,34 ; and model different clinical scenarios.^35–37 

The use of the FEA method in implant biomechanics analysis offers many advantages over other methods in simulating the complexity of clinical situations. On the other hand, this method is sensitive to several assumptions as they depend on many individual factors, including material properties, boundary conditions, interface definition, and also on the overall approach to the model.^38–42 

Also, strain gauge analysis was used by several examiners to investigate the biomechanical loading situation of implants during biting actions, yet it was mentioned that it is capable of detecting strains in only a limited sector of the peri-implant area.^43–45 

Thus, this study aimed to evaluate the biomechanical response of the peri-implant bone to standard, short-wide, and mini implants supporting single crowns under 2 different loading conditions, using strain gauge and FEA. The rationale for the investigated parameters was based on the idea of predicting the best treatment option to achieve long-term success.

An implant of standard diameter and length (3.75 mm diameter, 13 mm length; Screw Plant, Legacy system, Implant Direct LLC, Calabasas Hills, Calif), a mini implant (3 mm diameter, 13 mm length; Screw Direct one piece, Spectra system, Implant Direct LLC), and a short-wide implant (5.7 mm diameter, 8 mm length; Screw Plant, Legacy system, Implant Direct LLC) were investigated under a standardized test setup. Abutments 4 mm high and of matching width (Legacy system, Implant Direct LLC) were fixed and tightened to the internally hexed implants. The implant-abutment units were anchored in an epoxy resin block using autopolymerizing polyester resin (polypoxy 700, polymer, Chemical Industries for Construction Co, CIC, Egypt) in a straight-line configuration using a paralleling device (Bego fixator, Bego, Germany). The epoxy resin was used to simulate bone matrix, as it has mechanical properties similar to those of trabecular bone (ie, Young's modulus equals 3000 MPa).²¹  The implants' parallelism was checked by the aid of a surveyor (Ramses, Egypt). These implants were used to support single crown restorations.

Three full metal crowns made of nickel-chromium alloy were constructed to be supported by the implants. Crowns were constructed with standardized dimensions simulating mandibular second molars, with nonanatomical occlusal surfaces. Each crown was cemented to its corresponding implant-abutment assembly.

Each implant received 4 strain gauges (Tokyo Sokki Kenkyujo Co, Ltd, Tokyo, Japan) placed on the mesial, distal, buccal, and lingual surfaces of the epoxy resin adjacent to the implants. At these selected sites, the thickness of the acrylic resin surrounding each implant was reduced to approximately 1 mm and was adjusted to be parallel to the long axis of the implant-abutment assembly. Active strain gauges were bonded to their corresponding sites using cyanoacrylate adhesive. The lead wire from each active strain gauge was connected to a multichannel strain meter to register dynamic resin model microstrains transmitted to each strain gauge.

After temporary cementation (TempBond, Kerr, Orange, Calif) of each crown on its corresponding implant abutment, all strain gauges were set to 0.

Functional loads of 300 N were applied to the crowns using a universal testing machine (model LRX-plus; LIoyd Instrument Ltd, Fareham, UK). The machine is computer controlled by the Nexegen version 4.3 software, which permits the collection of the data. Two types of static axial loads were applied with 0.5 mm/min speed; the first load was 300 N, applied axially in the position of the centric fossa of each crown, while the second load was 3 mm off-axial distal loading of 300 N (Figure 1).

The B/L and M/D strains were recorded separately for each strain gauge. All recordings were repeated 3 times, allowing the strain indicator to recover to 0 strain before reloading. The 3 recordings were averaged, and the range of recordings was noted to assess the reliability of the recording system.

Electric strain gauges used were 2 mm in length, with a 2.11% ± 1% gauge factor, and a 119.8 ± 0.5Ω resistance. They use the property of the resistive element, which changes its electric resistance when strained. Thus, they were used to measure the strain induced around implants after load application, by measuring the change in their resistance and then calculating the amount of strain at the site of their attachment using the following equation: GF = ΔR/R over ΔL/L or GF = ΔR/R over E, where R is gauge resistance, ΔR is change in resistance during elongation, L is initial gauge length, ΔL is change in length, and E is the strain being measured.^47,48 

The primary focus of this study was the change in the magnitude of strain as the implant type was changed under different loading conditions.

Data were presented as means and standard deviation (SD) values. One-way analysis of variance (ANOVA) was used for comparison between the 3 implants. The Tukey post hoc test was used for pairwise comparison between the means when the ANOVA test was significant. A paired t test was used to compare between-axial and off-axial loads on each implant type.

The significance level was set at P ≤ .05. Statistical analysis was performed with SPSS 16.0 (Statistical Package for Scientific Studies) for Windows.

The current FEA aimed to simulate a clinical situation for different types of implants inserted in the position of a missing second mandibular molar to support a single crown.

Modeling of each implant using the SolidWorks 2007 software computer program was made according to determined measures. Modeling of the mandible was done by representing both compact and cancellous bone in a form of bone block; at the superior and inferior surfaces, a cortical bone layer with 2-mm thickness was simulated. All other bone modeling was done to simulate cancellous bone surrounded by compact bone. Different types of titanium implants and abutments were modeled as a unit. The embedded part of the implants corresponded to the length and diameter of each implant, while the 4-mm-high conical structure at the top simulated the abutment with a 9° convergence angle. NiCr crowns 6.5 mm high, 7 mm buccolingual, and 8 mm mesiodistal width with flat occlusal surface were created on top of the abutments with cervical dimensions corresponding to the diameter of implant fixtures (Figure 2). Cosmos Works2007 SP0.0 software, which is an FEA program, was used for analysis of models.

Assumptions were made that all materials in the FEA model were homogenous, isotropic, and linearly elastic. Moduli of elasticity and Poisson's ratios were used in the modeling of both compact and cancellous bone, implants, and nickel-chromium alloy of crowns (Table 1).^49–54 

All components were constructed in a way that ensures 100% contact along interfaces with no gaps or interferences.

A bonded contact was defined for every 2 contacting surfaces along the interface, which means that these objects are displaced as 1 unit upon load application and that the 2 contacting bodies cannot be separated or penetrated. Thus, the implant was rigidly anchored in the bone model along the entire interface. The same type of contact was provided at the crown-abutment interface.

A high-quality solid mesh was used in this study to create 3-dimensional parabolic tetrahedral solid elements.

Because a solid mesh was used, the resultant nodes were allowed to translate along any of the 3 orthogonal directions unless a restraint was applied, but no rotation was allowed.

The global average element size was set to 0.6 mm. The Jacobian check for solids was set at 4.

Two different loads were applied for each model. Each load was applied on a separate study.

Axial vertical load, 300 N in magnitude, was applied on the central fossae of the metal crowns.

Three-millimeter off-axial vertical load, 300 N in magnitude, was applied on the distal marginal ridge of the metal crowns.

The restraint property allows description of displacement on vertices, edges, or faces for use with the static analysis.

For the different studies, all of the elements were allowed to translate in all directions. The only restraint applied was a fixed restraint on the inferior surface of the mandible (the bottom surface), so no translation was allowed for this surface in all directions.

The analysis was run by an iterative method to solve the equations using appropriate techniques whereby, in each iteration, a solution is assumed and the associated errors are evaluated. The iterations continue until the errors reach an acceptable level.

The data were collected and tabulated to compare the reaction stresses, displacement, and strain of each model in response to the same loading.

In axial loading, the highest tensile microstrains were recorded in the area adjacent to the buccal surface of the implant (+574 ± 12.3 μm), and the highest compressive microstrains were recorded in the area adjacent to the distal surface of the implant (−889 ± 17.3 μm).

In off-axial loading, the highest tensile microstrains were recorded in the area adjacent to the mesial surface of the implant (+387.5 ± 14.1 μm), while the highest compressive microstrains were recorded at the area of load application distally (−1939 ± 22.8 μm).

In axial loading, recorded values of microstrains were significantly higher than those recorded in conjunction with the standard implant. The highest tensile microstrains were recorded in the area adjacent to the lingual surface of the implant (+650 ± 21.4 μm), while the highest compressive microstrains were recorded in the area adjacent to the mesial surface of the implant (−1626 ± 22.4 μm).

Off-axial loading increased the recorded microstrains 2- to 3-fold compared with axial loading. The highest tensile microstrains were recorded in the area adjacent to the mesial surface of the implant (+1779.8 ± 17.7 μm), and the highest compressive microstrains were recorded at the area of load application distally (−3631 ± 14.8 μm).

In axial loading, recorded values of microstrains were significantly higher than those recorded in conjunction with standard implant and with mini implant. The highest tensile microstrains were recorded in the area adjacent to the lingual surface of the implant (+1095.7 ± 11.2 μm), and the highest compressive microstrains were recorded in the area adjacent to the distal surface of the implant (−1205.7 ± 20.1 μm).

In off-axial loading, the highest tensile microstrains were recorded in the area adjacent to the mesial surface of the implant (+660.3 ± 20.3 μm), while the highest compressive microstrains were recorded at the area of load application distally (−1401.6 ± 20.1 μm).

The results of the ANOVA test for comparison between the 3 implants (Table 3) showed that in axial loading, the statistically significantly highest mean values of microstrains were recorded for the mini implant followed by the short implant and then the standard implant, for which the lowest mean values were recorded. In off-axial loading, the statistically significantly highest mean values of microstrains were recorded for the mini implant followed by the standard implant and then the short implant, for which the lowest mean values were recorded.

It should be noted that the mean microstrain values with off-axial loading were statistically significantly higher than with axial loading for standard implants, mini implants, and short-wide implants, as shown in Table 4.

Stress patterns are shown as contour lines with different tones connecting equivalent stress points between certain ranges. Figures 3–6 show the stress distributions in bone with different implant types and different loading conditions. All results of the numerical analysis are shown for Von Mises stresses and strains (Table 5).

All 3 types of implants showed similar stress distribution in bone. In all models under axial loading, the highest Von Mises stresses were observed at the implant bottom or base (corresponding to the cancellous bone; Figure 3), whereas under off-axial loading, the maximum Von Mises stresses were concentrated in the cervical regions at the upper edge of cortical bone around the implant neck, just under the area of load application (Figures 4–6).

The narrow-diameter implant (mini implant) generated the largest maximum Von Mises stresses, followed by the short-wide diameter implant, and the standard implant, which generated the smallest maximum Von Mises stresses. The stresses were more pronounced with off-axial loading than with axial loading.

The strain values recorded were obviously the highest for the narrow-diameter implant (mini implant), followed by the short implant then the standard implant, which showed the lowest strain values.

Moreover, the type of loading had a significant effect on the strain magnitude, as the magnitude of maximum Von Mises stresses significantly increased with off-axial loading compared with axial loading.

However, comparing all implant types and loading conditions, the experimental strain gauge analysis and FEA-simulated results did show a consistent relationship between the qualification of induced stresses and strains. Yet there was inconsistency regarding strain gauge analysis and FEA results between standard and short implants in off-axial loading, as the standard implant always had the lowest strains under axial and off-axial loading using FEA simulation. While strain gauge analysis showed consistent results in axial loading, the short-wide implant showed the lowest strains under nonaxial loading.

In addition, strain gauge analysis showed lower strains than the FEA method in all tested conditions, except off-axial loading of the mini implant, which showed significantly higher strain values.

To achieve stable osseointegration for implant restorations, the generation of high-stress concentration in bone should be avoided by maximizing the stress distribution, since it can induce severe resorption in the surrounding bone, leading to gradual loosening and finally complete loss of the implant.⁵⁵  Therefore, in this study, the effect of external forces on bone overload was investigated for different implant dimensions using strain gauges and FEA. The maximum normal stress criterion (ie, 3000 μ€) was used to evaluate the extent of the regions in which the normal stresses were beyond the allowable tensile and compressive values in the cortical bone.

It was found that the range of occlusal forces in second premolar and molar teeth is 210–400 and 130–395 N, respectively.^55,56  Thus, a 300 N load was applied on the tested implants supporting single crowns to be a clinically relevant load value.

Screw-vent implants were used because threads of implants decompose the axial load into 2 components: parallel and perpendicular to the plane of the threads. Distribution of the same force over a larger surface leads to lowering of the stresses. Therefore, the screw-type implant was chosen to reduce the risk of overloading of the surrounding bone.⁵⁸ 

The rationale for applying the loads on flat occlusal surfaces was to compare axial loading with absolute off-axial loading. As in the presence of cusp inclination, an additional horizontal load would be applied depending on the amount of cusp inclination, thus leading to reduction of the amount of vertical load transferred to the implants.⁵⁹ 

During clinical loading of an implant, the direction of loading is rarely along its central long axis, so the applied occlusal force is frequently in a direction that creates a lever arm, causing off-axial and bending moments in bone.⁶⁰  Thus, by measuring axial and off-axial loads, it was possible to evaluate the load transfer characteristics not only under regular masticatory forces but also under extreme load levels, such as those that occur during parafunction.

Based on the mechanostat theory stated by Frost, bone-remodeling activities remain in equilibrium when bone strain is in the range of 200–1000 μm. A high strain level ranging from 1000–3000 μm stimulates remodeling activity and results in an increase in bone density. A bony structure subjected to pathologic strain greater than 3000 μm induces generation of internal cracks that cannot be repaired by normal remodeling activity and will cause bone failure.³⁰  Thus, this criterion was used, which provides a method to identify failure regions due to tensile or compressive overloading.

Regardless of the method of testing, all loading conditions with different types of implants were within the normal physiologic loading zone of bone, 1000–3000 μm, except in the strain gauge analysis results of off-axial loading of mini implants, which showed microstrain values exceeding 3000 μm. The FEA of mini implants under off-axial loading similarly showed the highest microstrain values, yet they were within the normal physiologic loading zone of bone (Table 5).

Regarding the direction of loading, it was shown that changing the position of occlusal loading had a considerable effect on the amount and distribution of stresses. This was confirmed by FEA (Figure 3–6), which revealed that axial loading generated an even distribution of load around the implant in comparison to off-axial loading, in which stresses were more pronounced in the area of load application.

This may be explained by the increase of the horizontal component of the applied load, which generated an increase in the moment and eventually an increase in the compressive load on the side of applied force, to a level higher than the compressive load generated by only the vertical component of the force, generated by axial loading. This was in agreement with many authors who expressed their concerns that horizontal forces directed to implants may contribute to bone resorption or angular defects,^62–66  thus emphasizing that absolute axial loading on implants is required because bending overload may induce bone resorption around the implant collar and cause its failure.^67–69  This can be provided by using narrow occlusal surfaces and occlusal contacts that provide axial loading and by correct implant positioning, together with the selection of proper implant diameter.^70–72 

Furthermore, the maximum stresses and strains in off-axial loading in the jaw bone appeared mainly at the upper edge of the cortical bone near the neck of the implant and in the side where the force was applied. This corresponds with clinical observations, which reported that marginal bone loss always occurs around the necks of implants, which could be detrimental to loading conditions by increasing the lever-arm effect and bending moments on the implant.^73,74 

Considering the effect of implant diameter, using mini implants resulted in the highest stresses and strains both in axial and off-axial loading, when compared with both standard and short-wide implants. This might be attributed to the smaller surface area and volume of mini implants, which places more force per square millimeter against the encasing bone than larger-diameter implants.⁷²  This is because for every 0.5-mm increase in width, there is an increased surface area between 10% and 15% for a narrow range of diameters, and the percentage change is greater for smaller diameters and lesser for larger diameters.⁷⁵  Thus, it was concluded by Kong et al⁷⁶  that the increase in width of the implant may decrease stress by increasing the surface area, which may also reduce the length requirement.

Short-wide implants resulted in lower microstrains than mini implants both in axial and off-axial loading. The reduced strains associated with wider implants may be due to the increased structural capacity and the enlarged bone-implant contact area offered by these implants, resulting in a lower torque effect in conjunction with off-axial loading. In accordance, Belshi et al³⁸  indicated that a molar crown supported by a standard or narrow-size implant can easily introduce large bending moments to bone because the dimensions of the crown are usually greater than the diameter of the implants. Thus, the wide implant is suggested for placement at the molar region to reduce the possibility of overloading.^73,74,77 

Nevertheless, contradicting clinical studies demonstrated that the success rate of an implant was proportional to the implant length,⁷⁷  which was attributed to the decrease in the dissipation of occlusal forces due to their decreased length.⁷⁶ 

Strain gauge analysis and FEA methods showed comparable results in the qualification of strains induced under different conditions, except results of off-axial loading of standard and short-wide implants. Yet there was a discrepancy in the quantification of the strains due to the fundamental differences between the 2 techniques. The reason might be due to the difference in the modulus of elasticity of the acrylic resin model used to attach the strain gauges, in comparison with the modulus of elasticity of compact and cancellous bone used in the FEA method. Other reasons are that perfect bonding or connection between an implant and an abutment assumed in FEA is not the actual scenario for dental implants used in strain gauge analysis, in addition to the lack of preload application in the finite element model.⁷⁸ 

There are limitations associated with the current strain gauge analysis and FEA study. First, the bone-implant interface was assumed to have 100% osseointegration. If lower levels of osseointegration were to be used in the model, the bone strain levels would be expected to rise, as the same external force would be distributed to the bone through a smaller interfacial area.⁵⁹  Second, only 1 off-axial force was applied to the models. In reality, the occlusal force can be multidirectional, which may complicate the stress situations. However, it has been demonstrated in this study, through using axial and off-axial loading, that the trend in stress comparison between different designs remains the same. With a linear, elastic model used in this study, it is unlikely that other loading conditions will yield a different result with regard to different implant designs. Third, the material properties of the mandible were assumed to be regionally homogenous, which may primarily affect the resultant stress values.⁷⁹ 

Within the limitations of this study, the following conclusions were drawn:

• •

  Standard and short-wide implants could be a better choice than narrow (mini) implants in supporting single-unit restorations.

• •

  Off-axial loading not only increased values of maximum stress and strain compared with axial loading but also worsened the stress/strain distribution patterns in the bone-implant system.

• •

  Overloading occurs near the superior region of compact bone, and it is primarily caused by the off-axial loading.

Abbreviations

ANOVA

analysis of variance

FEA

finite element analysis