1.

Introduction

Chairside computer-aided designing/computer-aided manufacturing (CAD/CAM) systems are a reality in modern dentistry and the virtual model can be obtained with different methods. The crowns are manufactured based on the generated digital model by optical scanning of the mouth, regular impression (RI), or master cast (MC), which is directly controlled by the operator. After scanning, the crown is designed with a specific software using the virtual model, and a ceramic block is milled.

Historically, fixed prostheses are fabricated on MC made from a dental stone after the RI procedure. CAD/CAM equipment allows for obtaining the final restoration from an optical scanning directly from the mouth or from the MC in the laboratory. Therefore, most software allows its users to create a virtual model from the RI scanning without the need to obtain an MC.

The scanner CEREC 3D Bluecam (Sirona Dental Systems GmbH, Bensheim, Germany) allows one to obtain the virtual model from the three methods reported above and requires the intraoral powder application to scan directly in the mouth. The technical principle of the CEREC Bluecam scanner is that it projects a light stripe pattern on the object. As each light ray is reflected back to the sensor, the distance between the projected ray and reflected ray is measured. Because the fixed angle between the projector and sensor is known, the distance to the object can be calculated through the Pythagoras theorem, as one side and one angle (the fixed angle) of the triangle are now known.¹ The aerosol powder has to cover the prepared tooth, adjacent tooth, and mucosa, with a thickness sufficient to create an optimum image quality with a uniformly reflective surface for better scanning. Although there is the possibility to obtain the virtual model from all three methods, it is important to know if one has an advantage over the others regarding the marginal fit of the final crown.

The measurement of the marginal gap, which is a perpendicular measurement from the internal surface of the crown to the margin of the prepared tooth,² has been reported in the literature to establish a comparison of methods to obtain the virtual model.³ Precise marginal fit of all-ceramic crowns is vital for long-term success because cement is the weak restorative link.⁴ When cement dissolution occurs,⁵ a fissure is established between the dentin and crown.

Different techniques have been chosen to evaluate the marginal fit of crowns. Most demand a transversal section of crown/tooth to measure the gap.^6–8 A new technique that uses micro-computed tomography (micro-CT) has been used to investigate the marginal gap and has the advantage of being nondestructive.^9–11 This technique allows the three-dimensional investigation of small objects with high resolution. The marginal gap is obtained within the range of a few micrometers at multiple sites and in multiple directions.¹² Furthermore, even very proximate sections are possible.¹¹

Therefore, the current study evaluated the different methods for obtaining the virtual model with the chairside CEREC CAD/CAM system, using micro-CT to compare the marginal gap of feldspathic ceramic crowns. The null hypothesis was that different methods for obtaining a virtual model with the chairside CEREC CAD/CAM system would be similar.

2.

Methods and Materials

2.2.

Measuring Procedure

No internal adjustments, glazing, or polishing were made to the crowns before the marginal gap measurements to avoid any human interference. The prepared tooth was removed from the typodont and each crown was fitted and fixed. The fixing procedure was made for each crown at the same prepared tooth with a silicone material (Fit Checker, GC Dental Industrial Corp, Tokyo, Japan), by digital pressing, simulating a clinical situation (by a prosthetic specialist). To obtain images for marginal gap measurements, all samples were scanned using micro-CT (Micro-CT Scanco CT40, Scanco Medical AG, Zürich, Switzerland) at the Biological Research Imaging Center (University of North Carolina). Imaging was performed at 70 kVp and $112\mu \mathrm{A}$ with a resolution of $1024\times 1024\text{pixels}$. Pixel size and slice width were both $8\mu \mathrm{m}$ and the scan time was $\sim 1\mathrm{h}$. A total of $\sim 600$ two-dimensional images were acquired for each specimen. The transaxial images of the crown and prepared tooth were first obtained, as shown in Fig. 1.

Fig. 1

Micro-computed tomography transaxial image of the crown and prepared tooth.

Thirteen images from the sagittal set [Fig. 2(a)] and 13 from the coronal set [Fig. 2(b)] were selected to show the sample extension in two dimensions. From the total amount of the obtained images, the authors selected 13 images that were equally distributed between the first and last image in which the cervical margins appear. In each image, two measurements for horizontal fir and two for vertical misfit were done at $400\times$ magnification using the CTAn processing software (Version 1.12.0.0, Skyscan, Kontich, Belgium).

Fig. 2

(a) Sagittal image. (b) Coronal image.

For vertical misfit, the measurements were made from the most external point at the preparation margin of the tooth to the crown margin [Fig. 3(a)]. For horizontal misfit, the measurements were made from the external crown margin to the most external point of the tooth [Fig. 3(b)].

Fig. 3

(a) Vertical misfit. (b) Horizontal misfit.

The marginal gaps were measured at 52 sites for each sample, following Groten et al.’s¹⁴ recommendations. In this way, 260 vertical misfit and 260 horizontal misfit measurements were done per group. Three independent examiners were involved in the measurements and the mean value was obtained. To ensure standardization among examiners, they were calibrated prior to the measurements.

3.

Results

The values of vertical marginal gap for each group are presented in Table 2.

Table 2

Mean and standard deviation of vertical misfit (μm).

The vertical marginal gap of VITABLOCS Mark II crowns made by RI protocol showed most of the measurements to be $>75\mu \mathrm{m}$ (50.8%); for the other groups, most of the measurements were $<75\mu \mathrm{m}$. Vertical misfit values $<75\mu \mathrm{m}$ were considered the ideal value in this study (Table 3).^2,15–19

Table 3

Marginal fit percentage for each group.

The horizontal measurement values were divided into the following categories: (1) underextension, (2) equally extended, and (3) overextension of the crowns fitted to the prepared finish line (Fig. 4).

Fig. 4

Percentage of underextension, equally extended or overextension of crowns fitted to the prepared finish line.

4.

Discussion

Although it is known that the smaller the marginal gap, the better the fit at the tooth/crown interface, there is no consensus about the acceptable value. Over the years, the criteria used range from 75 to $>120\mu \mathrm{m}$, as reported by several researchers.^2,15–19 For this study, because of the lack of consensus and the current technology available, the vertical gap values were categorized and grouped according to the different values presented in the literature.^2,15–19

The conditions to evaluate the restoration for clinical acceptability depends on several factors, such as method of evaluation (clinical), instrument (explorer probe), examiner ability, and analyzed materials, all of which can lead to different results. In vitro studies evaluated the marginal fit of ceramic crowns with 5 to 25 samples for each test group and 6 to 50 sites for each crown, resulting in 60 or more marginal gap values analyzed per group.^{3,7,8,10,20,21} Other studies performed only four measurements per sample,^2,15 which could not be representative of the marginal gap. This study evaluated five different crowns for each group fixed to the same tooth. Measurements were made at 52 sites in each crown, totaling 260 values for each group, in order to obtain clinically relevant information about the marginal gap (Table 2). The suggestion comes from a previous study¹⁴ that reports that at least 50 measurements well distributed in the margin of a sample are necessary to obtain clinically relevant information. The methodology used in the present study permitted a larger number of measurements per sample. With a greater number of measurements, more confidence and accuracy can be achieved.

The measurements were performed in images originated from a micro-CT. Such equipment has been used to investigate the marginal fit and has the advantage of being a nondestructive technique compared to others. The stereomicroscope requires a transverse section of the crown and tooth to measure an internal misfit^7,20,22 and it can cause deformations.⁷ Moreover, micro-CT allows the analysis in slices without any sample sections, so there is no overlap, unlike other techniques, such as scanning electron microscopy.^17,23 Measurements can occur at different distances along each section and the marginal zone, providing true three-dimensional reconstructions.¹⁰ In the present study, it was possible to perform measurements at $400\times$ magnification, facilitating the measurement procedure. In other studies, the maximum magnification described was $\sim 250\times$.^2,24 This method is a useful tool for the evaluation of the marginal fit of dental restorations.^10,11 However, micro-CT analysis can be expensive and time consuming.²⁵

Another important issue to be considered is how the crown is fixed on the teeth for micro-CT scanning. Recent studies have used occlusal loading or a constant torque control device to keep the crown on the tooth with no need for cementation.^3,20,21 Another study used a clinical cementation process.⁷ In this case, the sample is subjected to a transverse cutting method to measure its internal portion using a technique that can cause destructive deformation to the prosthesis.²⁶ The cementation process can damage the teeth and interfere with the results of the depth measurement.^9,27 Furthermore, the cement can increase the discrepancy of the marginal extent, since the excesses adhere to the tooth surface and to the restoration.²⁴ In our study, the same tooth was used with each crown and all the crowns were fixed by a prosthetic specialist with a silicone material, by digital pressing, simulating a clinical situation and standardizing the comparisons. It is noteworthy that after fixing, the crowns showed high resistance to being removed from the tooth, confirming that there was no movement during measurement.

Table 2 shows the percentage values related to the vertical misfit. The vast majority of problems related to failure in fixed prosthesis is caries due to microleakage at the cement line.²⁸ Thus, while the literature is still controversial,^2,15–19 the authors of the present study believe that the limit of acceptable misfit should be up to $75\mu \mathrm{m}$, because it was the lower value described in the literature¹⁵ for ceramic crowns.

Restorations with horizontal misfit facilitates food and bacteria retention,^29,30 making hygiene more difficult and leading to periodontal problems and cavities, reducing the longevity of rehabilitation. The authors handled the horizontal and the vertical misfit differently because it seems to have different clinical implications. When a horizontal misfit occurs, there is a possibility to make adjustments³¹ in the crown or in the teeth before the cementation. When the vertical misfit occurs, there are no adjustment possibilities, leading to an increased risk of periodontal disease, microleakage, and recurrent caries.^30,32 So the authors presented the vertical misfit as being divided into four categories according to the literature^2,15–19 and, for horizontal classification,³³ considered the overextension as negative because it seems to be the worst situation, increasing the potential for plaque accumulation.

The marginal fit in CAD/CAM is dependent on the size of the drill, precision of the milling unit, digital model obtention,^3,21 and also the machine calibration and system for image capturing. So to make a fair comparison, it is necessary to consider the system, its version, the measurement technique, the type of restoration (crowns, inlays, onlays), and restorative material.³ In this study, all machines were calibrated prior to the study and the parameters of the software were the same for all groups.

The groups scanned from MC showed good values for marginal fit. These MC all originate from the typodont RIs, suggesting that the RI was precise. The worst values found in the RI group should be related to a possible difficulty of scanning the surface of the impression material, even with the application of powder. The fact is that the groups T, MC, and powdered MC had a very similar behavior, inferring that restorations obtained by these three ways may achieve optimum fit, although direct scanning from the mouth provides restorations in a shorter time.

The position of the sample was not standardized into the micro-CT. In this way, it was impossible to correlate the marginal gap and at which site it occurred. In future studies, to standardize measurements, it would be interesting to use an index. Another interesting study is to variate the luting space parameters in the software to verify whether it interferes with the marginal gap of all-ceramic crowns obtained by CAD/CAM. Moreover, long-term clinical data are necessary to verify the efficacy and importance of these techniques.

5.

Conclusions

Based on the results of this study, it was concluded that the virtual model obtained by scanning the mouth or the MC, with or without powder, showed acceptable values of marginal gap on feldspathic ceramic crowns (VITABLOCS Mark II) manufactured with a chairside CEREC CAD/CAM system, considering the ideal vertical misfit up to $75\mu \mathrm{m}$. The higher values of the marginal gap for scanning the regular impression suggest that it is preferable to directly scan the mouth or the MC.