**3. Results**

Descriptive statistics are given as means with 95% confidence intervals (CIs) for the variables of maximum total time, average total time, average load, and maximum load for each group (Table 2).

**Table 2.** Means with 95% confidence intervals for each factor and differences in clinical factors by the group.


Max.: maximum; CI, confidence interval; Group 1 (CCNPPCs), Group 2 (CMTPCs), Group 3 (CPTPCs), Group 4 (CMZPCs); \* Games–Howell test; \*\* Tukey's adjusted post hoc test; # *p*-value significant at *p* < 0.05.

Using the one-way ANOVA with Tukey's adjustment, significant differences in the average-load variables were observed between the four different groups, with higher values recorded for Group 3 (CPTPC) in comparison with all other groups (Figure 8).

The mean difference values are shown in Table 2. The load values necessary to create failure in Group 4 (CMZPCs) were lower when compared with those of Group 3 (CPTPCs), Group 2 (CMTPCs), and Group 1 (CCNPPCs); furthermore, all groups showed different distributions at the time of failure and were statistically significantly higher for Group 3 (CPTPCs) and Group 2 (CMTPCs) (*p* < 0.05). Group 3 (CPTPCs) and Group 1 (CCNPPC group) required loads of 361.5 N and 295.9 N, respectively, to fail, and the difference was statistically significant (*p* < 0.05). When considering Group 4 (CMZPC group) and Group 3 (CPTPCs), they required loads of 361.5 N and 248.1 N, respectively, and this difference was statistically significant (*p* < 0.05). The differences in the averages of maximum time and average total time between Group 3 (CPTPCs), Group 2 (CMTPCs), Group 1 (CC-NPPCs), and Group 4 (CMZPCs) were significant (*p* < 0.05), with higher values recorded for Group 3 (CPTPCs) in comparison with all other groups (Figure 8).

Using the Mann–Whitney U test with Bonferroni adjustment, no significant differences in the medians of the displacement scores were observed between each group and the control group (Group 1), as shown in Table 3. All Bonferroni adjusted *p*-values were >0.05.


**Table 3.** Comparison of different post-and-core fabrication materials with the control group.

*p*-value significant at *p* < 0.05.

The failure mode varied between the groups, according to the scanning-electronmicroscopy images. The majority of failures for all tested custom post-and-core systems were cohesive failures, except for Group 4, which revealed a mixed type of failure, as shown in Figures 9–12.

**Figure 9.** Scanning-electron-microscope image of (**A**) Sectioned root canal, and (**B**) displaced post, at 200× magnification, showing cohesive failure mode in Group 1 (CCNPPCs) sample at root/post interface.

**Figure 10.** Scanning-electron-microscope image of (**A**) Sectioned root canal, and (**B**) displaced post, at 200× magnification, showing cohesive failure mode in Group 2 (CMTPCs) sample at root/post interface.

**Figure 11.** Scanning-electron-microscope image of (**A**) Sectioned root canal, and (**B**) displaced post, at 200× magnification, showing cohesive failure mode in Group 3 (CPTPCs) sample at root/post interface.

**Figure 12.** Scanning-electron-microscope image of (**A**) Sectioned root canal, and (**B**) displaced post, at 200× magnification, showing cohesive failure mode in Group 4 (CMZPCs) sample at root/post interface.
