Figure 6 and Table 2 show the results for implant displacement under 100 N of vertical loading at each loading point and in each model.

The implant displacement under loading conditions in the experimental model and the two FEA models showed a tendency to exhibit the smallest values under central loading; substantially similar values were exhibited in buccal and lingual loading. Buccal loading (p < 0.05) in the experimental model and buccal (p < 0.01) and lingual loading (p < 0.05) values in the FEA models were significantly greater than the values obtained from central loading. The implant displacement under loading conditions in the FEA models showed lower values than in the experimental model at all loading points, but aspects of implant displacement under loading caused by differences in the loading point showed a similar tendency. The correlation coefficient between the experimental model and the contact model was 0.925, representing a significant and strong correlation (p < 0.01). The maximum CV value was 4.90% in the experimental model, 9.64% in the contact model, and 9.26% in the fixation model (Table 2).

Figure 7 shows the results of three-dimensional implant displacement for each loading point under 100 N of vertical loading.

Under buccal and lingual loading conditions, displacement involving rotation inclined towards the loaded side was exhibited; the displacements were substantially equal (Figure 8). Central loading resulted in the smallest displacement, and almost no displacement was observed. The fixation model had less displacement than the contact model. With regard to the aspects of displacement, similar tendencies were shown in both the contact model and the fixation model. The results of the ANOVA showed that for both the contact and fixation models, the loading site was a significant factor in the three-dimensional displacement (p < 0.01) (Table 3).

At all three loading points, no. 34 and no. 35 showed displacements that were rotated and inclined towards the distal direction; in contrast, no. 36 showed a displacement that moved parallel to the distal direction (Figure 9). Compared with the contact model, the fixation model had less displacement, but aspects of the displacements showed similar tendencies. The results of the ANOVA showed that significant factors for three-dimensional displacement were assessment site, dental formula, and loading point in the contact model, and assessment site and dental formula in the fixation model (p < 0.05) (Table 4).