Considering the difficulty of in vivo investigation, simulation studies are valuable in investigating the load distribution of IARPD, although the number of studies is limited. This review summarized the current biomechanical findings regarding the load distribution of IARPD from in vitro model experiments and FEA studies. These studies included various biomechanical aspects of IARPD; however, the review focused only on implant location’s effects on load distribution in the mandibular free-end missing.

It was difficult to determine the better method for elucidating the biomechanics of IARPD, considering the advantages and disadvantages of each study design. Simulation studies should ideally use models reproducing the details of human jawbone morphology or properties and applying the real loading conditions. For example, 3-D FEA is generally superior to 2-D FEA. FEA can be more effective in investigating the stress/strain distribution in the jaw bone. However, in vitro model experiments might be able to make more sense of denture behavior. In most FEA studies, the clasp on the abutment tooth completely adhered to the tooth, which is not observed in the clinical situation. Understanding the characteristics of each simulation study before interpreting the results is essential.

In the in vitro model studies, it is challenging to imitate living tissues, such as the jawbone, mucosa, and teeth. Although researchers used an artificial mucous membrane and PDL using silicon materials in the model studies, the thickness and elasticity substantially vary among individuals in vivo. Similar to the model studies, it is unclear if the material properties of jawbones, PDL, and other components used in FEA studies were biologically relevant. Recent FEA studies have generalized nonlinear heterogeneous material properties based on a specific patient’s CT data, enabling more realistic simulation research [29, 30].

For the denture design of the IARPD, all studies used metallic frames. Previous FEA studies have demonstrated that the occlusal rest position or attachment system affects the strain on the metallic frame of the IARPD [31, 32]. Nogawa et al. [33] also compared the biomechanical behavior of three types of direct retainers of IARPD; however, further studies are still required to consider the retainers’ effect on load distribution. Elsyad et al. [19] compared the number of free-end missing teeth and clarified that the long saddle of IARPD recorded significantly higher peri-implant stresses than the short saddle. Further studies are needed to clarify the effect of the number of missing teeth on IARPD behavior.

In terms of loading conditions, a static load was applied to the IARPD in both in vitro and FEA studies. The magnitude of the applied load ranged between 50 and 200 N, thus simulating an occlusal force during clenching or chewing. Although only vertical load was applied in the model studies, oblique or horizontal direction loads were additionally applied to the FEA. However, in clinical scenarios, various directional dynamic loads are exerted on the tooth and implant during chewing [34,35,36]. The model study applying dynamic and static loading conditions to the denture demonstrated significant differences in the load distribution between loading conditions in the mandibular implant-supported overdenture [37]; therefore, the dynamic loading condition should be included in the simulation studies.

To understand the load distribution of the IARPD, the loads applied to each supporting element (abutment tooth, implant, and mucosa under the denture base) of the IARPD should be ideally measured three-dimensionally, simultaneously, and accurately. Strain gauges and piezoelectric transducers were mostly used to measure the load or stress of the supporting elements in model studies. In the included studies, the strain gauges were attached directly to the implant body [18,19,20] or the resin part around the implant [21, 22]. Considering the load in the peri-implant bone, the latter might be more meaningful, because the distortion of the surrounding bone can be more related to bone damage or remodeling. The piezoelectric transducer method can effectively and accurately measure the 3-D load on the implants and abutment tooth because of its favorable characteristic of load measurement in vivo [38]. For measuring the load under the denture base, Matsudate et al. used seat-type sensors [17], which were also used in vivo previously [39, 40]. However, a thin seat-type sensor with a larger sensing area may be ideal for understanding the load distribution in this area. Alternatively, FEA studies can evaluate the magnitude and distribution of the stress/strain in the bone, mucosa, and PDL. Although there are no explicit guidelines regarding the kind of stresses that should be used in the FEA for dental biomechanics, principal stresses and von Mises stresses are often used equally. Since minimum principal stress represents the peak compressive stress, the evaluation of that stress value could provide valuable information for understanding bone remodeling [41]. In the FEA studies included in this review, Ohyama et al. used the minimum principal stress for evaluating the distribution of mechanical stimulation in the model [27]. Since the accuracy and clinical validity of the FEA results are highly dependent on the reproducibility and condition settings of the model, more recent studies may be generally reliable due to the development of computational technologies. On the other hand, the FEA studies in this review have not been verified using clinical outcomes. Therefore, although the usefulness of FEA is understood, the clinical validity of such simulation results might not be high. This means that clinical validity must be carefully considered when interpreting FEA results, even in model experiments.

Regarding load distribution, the loads applied on the abutment tooth, the implant, the mucosa under the denture base, and their balance were considered. With regard to the load on the abutment tooth, the load magnitude and direction, as well as the stress on the PDL and the surrounding bone should be considered. Model studies revealed that the load on the abutment tooth increased when implants were placed in the second molar region, and the bending moment became larger when implants were placed in the premolar region. In particular, Matusdate et al. demonstrated a larger load on the abutment tooth in the IARPD with the implant location at the second molar compared with CRPD [17], which means that the implant placement does not necessarily reduce the burden on the abutment tooth in IARPD. When focusing on the stress in PDL or bone around the abutment tooth, the stress can be larger in the implant location at the second molar region than in other regions from the model experiment results or FEA [25, 27].

On the contrary, some studies showed that placing the implant closer to the abutment tooth caused more strain on the abutment tooth [23, 24, 28]. However, considering the contour diagrams of FEA results, the higher stress area was larger in the implant at the second molar region than in other regions in the above studies [23, 28]. This can be explained by the fact that the denture can rotate on the implant as a fulcrum, which may reduce the load’s vertical components but increase the load’s lateral components on the abutment tooth [17], causing more strain on the abutment tooth. Denture rotational movement can also be affected by the loading condition, namely, whether the loading point on the denture is anterior or posterior to the implant location [20]. In addition, Ohyama et al. suggested that denture and abutment tooth movement can be controlled by the bracing effect of the implant abutment [27]. If the denture behavior can be controlled well by the implants placed at the premolar area, the burden on the remaining teeth may reduce, protecting the remaining teeth. The survival rates of abutment teeth used to retain and/or support the IARPD were reported to range from 79.2 to 100% [42], which might be better than that (73.6%) of the abutment tooth in conventional RPD [43]. The implant support and/or retention in IARPD can avoid the swing movements along the axis of rotation of the prosthesis, which may reduce the risk of abutment tooth loss. Appropriate oral hygiene and a regular control and maintenance program are also essential to reduce the risk of failure of abutment teeth [42].

Considering the overall stress distribution in the mucosa area, the implant at the first molar area may minimize the total stress in that area [23,24,25]. Some studies showed that placing the implant under the denture base reduced the load on the mucosa [17, 28]. The previous model experiments [44, 45] also demonstrated minimized mucosal pressure upon placing the implant in the second molar area. It is to be noted that, as described above, even if the implant location was the same, the load under the denture base can change depending on the loading points on the denture [20]. When the loading point is set between the implant and the abutment tooth, the load on the mucosa can be significantly reduced. It may be reasonable to consider the main occluding areas [46] for each IARPD patient to determine the most optimum implant location.

Most studies showed that the load on the implant became larger in the implant location of the second molar area [17, 19, 22]. Other studies demonstrated that the bending moment of the implant [20] or peri-implant bone strain was larger in the implant location in the premolar location [18, 21]. Considering the stress in the entire jawbone, placing implants at the first molar region might be less stressful [23,24,25] and enhance balance [26]. However, the included FEA studies did not focus on the region of interest in the peri-implant bone for stress distribution. It is to be noted that the effect of bracing and retention of implant abutment can change denture behavior, affecting the load distribution in IARPD [22, 27]. Despite favorable clinical outcomes of the implants in IARPD [3, 5], there may be some concerns about peri-implant bone resorption; therefore, researchers should consider the burden on the implant in IARPD. On the other hand, defining an appropriate load distribution is difficult. The risk of mechanical or biological complications is thought to increase if the load is concentrated on any one supporting element in IARPD. Therefore, appropriate load distribution can be considered a state in which stress is not concentrated on any one supporting element.

Although the experimental studies included in this review reported the absolute values of load or mechanical stress on the supporting elements, they used them to assess the experimental conditions in each study. Thus, comparing the absolute values among the different studies was less meaningful. In addition, the effect of implant location on load distribution differed across studies, which may be attributed to the heterogeneity of the methodology used in these studies. In particular, model setting, loading condition, load or stress measuring methods, or assessment places were different. Due to the above limitations, the results were not analyzed statistically in this review. In addition, simulation studies warrant verifying the validity of simulation results with actual clinical data [47]. Although this review included the studies with IARPDs for three or more teeth-free-end missing, patients with two teeth-free-end missing also visit the dental clinic. Actually, one study included the case of two teeth free-end missing for both the model experiment and FEA and investigated the mechanical stress on the abutment tooth and implant of IARPD [48]. A shortened dental arch (no prosthesis or only implant-supported fixed prosthesis at the first molar) or a fixed prosthesis with two implants may be clinically adopted rather than the IARPD in such cases, but it is necessary to investigate IARPD for two missing teeth in the future.

Summarizing the studies comparing three implant locations in IARPD for mandibular free-end missing: the first or second premolar, first molar, and second molar areas, the effect of implant location differed among the studies due to the differences in the measurement method, such as the load measurement method or position, and loading conditions.

Overall, clinical suggestions can be provided for each implant position in the case of one implant-assisted removable partial denture in mandibular free-end missing.

Premolar region The condition of the peri-implant bone should be evaluated carefully, because the lateral load on the implant can be relatively high due to the rotational movement of the denture with the implant as a fulcrum. It is recommended when the abutment tooth is periodontally compromised and an implant in the premolar region would reduce the forces on the abutment tooth.

First molar region Considering the balance of load distribution to all the support elements of the IARPD, implant placement here may offer a greater favorable distribution and dissipation of load and stress among the various supporting elements.

Second molar region The load on the mucosa under the denture base may be reduced. The condition of the abutment tooth should be considered, and equal load distribution to the remaining teeth might be essential to prevent load concentration on the abutment tooth. It is recommended when periodontal conditions of the abutment tooth are stable.