*3.2. Marginal Misfit*

Means and standard deviation of marginal misfit of CoCr specimens are presented in Table 2. Samples fabricated from the DLP-Cast technique demonstrated the highest mean marginal misfit (147.746 ± 30.306 μm), whereas the lowest marginal misfit was established by SLM-fabricated specimens (27.193 ± 8.519 μm). Moreover, ANOVA revealed that there was a statistically significant difference in mean marginal misfit among all investigated groups (*p* < 0.05). Individual intergroup comparison using Tukey Kramer post hoc test revealed that copings fabricated from the SLM technique displayed lower marginal misfit than DLP-Cast (*p* = 0.001) and CAD-CAM (88.943 ± 20.880 μm) (*p* = 0.001). However SLM copings showed higher but comparable misfits to Cast-LWT (47.861 ± 19.693 μm) samples (*p* > 0.05). The microCT images of the assessed samples are presented in Figure 3.


**Table 2.** Comparison of marginal misfit of the CoCr copings.

\* Statistical significant difference among groups shown in corresponding rows and columns using Tukey Kramer post hoc test; \$ Statistically significant difference using ANOVA.

**Figure 3.** MicroCT images of the assessed samples in groups (**A**) Cast/LWT, (**B**) CAD-CAM, (**C**) DLP-Cast, and (**D**) SLM.

Figures 4–7 present the correlation between Ra and marginal misfit in CAD-CAM, Cast-LWT, SLM, and DLP-Cast study samples respectively. It was observed that Ra influences the marginal misfit in CAD-CAM (81.7%), SLM (94.8%), and DLP Cast (98.6%) technique-fabricated copings. (*p* < 0.01). Whereas, copings which are fabricated from CAST-LWT technique did not display any significant effect of Ra on the marginal misfit on the specimens of this group (*p* = 0.435), as displayed in the correlation plot (Figure 5).

**Figure 4.** Showing a positive correlation between surface micro roughness and marginal misfit in CAD-CAM samples. R2 showed 81.7% variation in marginal misfit explained by surface micro roughness; *p*-value was less than 0.01, and therefore statistically significant.

**Figure 5.** Showing a positive correlation between surface micro roughness and marginal misfit in CAST-LWT samples. R<sup>2</sup> showed 7.8% variation in marginal misfit explained by surface micro roughness; *p*-value 0.435, and therefore statistically insignificant.

**Figure 6.** Showing a positive correlation between surface micro roughness and marginal misfit in SLM samples. R<sup>2</sup> showed 94.8% variation in marginal misfit explained by surface micro roughness; *p*-value was less than 0.01, and therefore statistically significant.

**Figure 7.** Showing a positive correlation between surface micro roughness and marginal misfit in DLP Cast samples. R<sup>2</sup> showed 98.6% variation in marginal misfit explained by surface micro roughness; *p*-value was less than 0.01, and therefore statistically significant.

#### **4. Discussion**

The present in vitro study was based on the hypothesis that there is no difference on Ra and marginal misfit of CoCr copings manufactured by conventional (Cast-LWT) and contemporary techniques (CAD-CAM, SLM, DLP-Cast). However, the existing study revealed that the additive technique of SLM showed lower misfit and high roughness. Whereas, the DLP-Cast specimen displayed higher internal misfit and lower Ra than the conventional technique. Therefore, the postulated hypothesis was rejected. These outcomes can be attributed to the number of steps, scanning and software limitations, and non-optimal parameters.

Contemporary CoCr alloys have gained popularity as compared with conventional gold alloys due to improved corrosion resistance and low cost [23]. The available literature showed multiple techniques, i.e., the direct-view measurement technique, the silicone replica technique, and cross-sectioning to measure the marginal misfit of fabricated copings [26]. The micro-computed tomography (CT), on the other hand, is comparatively an advanced method to assess the fit of indirect restoration through processing of scanned specimens slices, reconstructing the assembly by using software and gaging the misfit [27]. Similarly, in the present study the LPM was used to evaluate the Ra of CoCr copings. It is an optical system which is used to scan comparatively larger surface areas. It exhibits advantages over other techniques, i.e., determining surface characteristics such as the height of the largest profile projection and the depth of the largest profile depression [28].

In the present study it was found that SLM fabrication technique displayed lower marginal misfit than the other investigated groups. The better performance of the SLM technique is in line with the results of the study conducted by Fathi et al. [29]. Although, in the study by Fathi et al., assessments were performed using a silicone impression technique [29]. This can be explained by the fact that a lesser number of steps involved in the technique contributes to the precisely fitting copings as compared to other tested techniques. Moreover, copings fabricated from Cast-LWT displayed higher marginal misfit as compared with SLM copings. Multiple factors explain the increased misfit of Cast-LWT compared with SLM samples. Multiple steps are involved in the production of prosthesis through Cast-LWT and each step poses a risk of incorporating error thus compromising prosthetic fit [16]. Moreover, the accuracy of cast coping obtained through Cast-LWT depends on the accuracy of wax pattern and technical accuracy, i.e., wax composition, tank and block temperature, time specified for cooling of the wax pattern, and the firing temperature necessary to achieve desirable outcomes [30]. In addition, distortion of inlay casting wax, its shrinkage, and high investment expansion affects the precision of copings fabricated with Cast-LWT [31].

In the present study, DLP-Cast copings exhibited the highest marginal misfit compared with all other techniques. Marginal fit is highly dependent on the material properties utilized to fabricate copings in the 3D printer [32]. Moreover, resin used in the DLP-Cast technique undergoes polymerization shrinkage which generates stress, resulting in distortion of internal and marginal misfit [33]. In addition, the effect of scan spray when scanning the models cannot be overlooked [4]. These findings are in accordance with the study conducted by Kim at el., which used M-One printer, resulting in greater marginal misfit of DLP-Cast than other groups tested [25]. Kim et al. assessed the misfit using the weight of the silicone material and a digital microscope with sectioned specimens. Furthermore, it was also found that CAD-CAM displayed a significantly higher marginal misfit compared with Cast-LWT and SLM copings. This may be because correctness of internal and the marginal fit of copings produced through the subtractive technique depends on bur size. Any discrepancy in selection and size of burs relative to the size of coping results in compromised marginal properties [34]. Moreover, scanning and software system limitations related to finite resolution plausibly explicate the findings as these might result in margins with lower fit accuracy [35]. Many studies have compared the marginal or internal fit of coping and crowns fabricated from CAD-CAM and Cast-LWT [5,12,36]. Yet, it is challenging to conclude the findings of those studies due to variations in sample size, methods adopted to measure the marginal or internal gap, the type of cement used, and the CAD-CAM systems chosen. However, results of multiple previous studies are in line with the finding of the present study and displayed larger marginal or internal misfit for CAD-CAM fabricated prostheses than conventional techniques [12,37,38].

The influence of the different fabrication techniques on the Ra of CoCr copings have been addressed, proposing that Ra varies with the type of manufacturing technique [30]. Available literature also suggested that Ra values of any indirect prosthesis should be at least 0.2 μm [39]. Ra is suggested to influence retention of restorations, but its effect on the marginal fit is not clear. In the present study, it was found that all fabrication techniques

demonstrated Ra higher than the recommended threshold. Moreover, the difference in Ra values among different AM techniques, i.e., DLP-Cast and SLM, can be explained by different parameters adopted during fabrication [40]. It is suggested that the use of non-optimal parameters in any of the AM techniques result in porosity and an increase in the Ra of the prosthesis [41]. In the authors' opinion, decreased marginal misfit in SLM-fabricated copings may be due to the highest surface roughness obtained in this group. Similarly, surface roughness of copings fabricated from the CAD-CAM subtractive technique depends on the cutting speed and depth of the cutting [42]. Moreover, LWT copings displayed the lowest Ra score; this may be due to the favorable inlay wax surface properties along with strict adherence to manufacturer guidelines aiding low Ra [43].

The study showed that additive manufacturing methods like SLM and DLP-cast showed varying success in producing marginally accurate and smooth metal copings. The findings suggest that SLM copings have a good marginal fit and high roughness; however, DLP-Cast specimens display low roughness but a poor marginal fit. Therefore, both of these techniques need further development and can only be applied in limited clinical contexts. The findings of the study should be interpreted in light of the possible study limitations. In the present study, copings were manufactured under ideal circumstances and controlled conditions. In addition, the outcomes of the study are limited to the materials and techniques employed in the experiments. Therefore, to translate the findings of the present study into clinical recommendations, randomized clinical trials assessing the fit and adaptation of CoCr copings fabricated with additive manufacturing techniques are warranted.
