Novel Trends in Dental Color Match Using Different Shade Selection Methods: A Systematic Review and Meta-Analysis

Since color matching is considered a subjective procedure, accurate shade choice is often the most challenging stage of recreating the natural appearance of teeth. Furthermore, accurate determination of tooth color is imperative for the final outcome of dental restorations. The purpose of this research is to assess the accuracy of color match between diverse shade selection methods throughout a systematic review and meta-analysis. Two independent investigators (L.H. and R.B.) screened the literature in five electronic databases. Randomized controlled trials or in vitro papers studying the effect of using either digital shade selection or visual shade selection on the accuracy of color match were included. A total of 13 manuscripts comprised the meta-analysis. Color difference (ΔE) between restorations where the shade matching was performed by the conventional method was greater than those where the shade matching was performed by computerized methods (p = 0.007). According to the subgroup analysis, only the use of digital photographs for shade matching showed a reduction in the (ΔE) (p < 0.0001), while the use of a spectrophotometer has no advantages over the use of visual shade guide tabs (p = 0.57). On the other hand, global analysis showed that incorrect shade matching was higher when the conventional method using shade guide tabs was used (p < 0.001), irrespective of whether a spectrophotometer or a digital camera was used (p < 0.001). This study concluded that the use of digital photography and spectrophotometric measurements led to fewer color differences and less incorrect shade matching than conventional methods using color shade tabs.


Introduction
The fabrication of indirect restorations necessitates worthy interaction between clinicians and laboratory technicians. Furthermore, shade choice is the often the most challenging stage in the recreation of the natural appearance of teeth [1]. According to previous studies, differences among practitioners concerning shade matching for the same teeth occur over days [2,3]. In fact, researchers have considered color matching as a subjective procedure, reliant on several influences, such as light source, object, and observer [4]. While there is no process considered as the gold standard, different methods to assess color in dentistry exist, including visual and instrumental [5].
Color communication using shade guides is the most shared technique. However, this method is considered subjective, since it is induced by age, sex, experience of the observer, eye fatigue, and ambient light [6]. In this sense, instrumental methods have gained popularity; however, they are expensive and not always available to the dentist [2,7]. It is important to note that instrumental methods include spectrophotometers, scanners, cross-polarizing filters, digital cameras, and smartphones [2,8]. These devices consist of a detector, signal conditioner, and software that process the signal to make the data usable in a clinic or laboratory [9]. Spectrophotometric color measurements are capable of reliably quantifying the color of both extracted teeth and dental materials [10].
Digital photographs are a common interaction tool in dental workplaces [11]. Digital photographs are able to capture clear images of the tooth and have been progressively used to document the whole clinical procedure [2,12]. Predictably, digital photographs of natural teeth can show noteworthy color modifications when the optimal conditions of light illumination are not used [12].
On the other hand, there has been an increased awareness for aesthetically pleasing restorations among patients. Hence, it is a practitioner's responsibility to provide restorations which can adequately mimic the surrounding natural dentition [13]. As scientific evidence comparing differences in visual shade matching using different devices is scarce, this topic remains controversial among researchers. Therefore, the purpose of the present study is to systematically review the literature to compare the accuracy of color matching between different shade selection methods. The null hypothesis tested was that all different shade selection methods would have the same accuracy.

Materials and Methods
This systematic review and meta-analysis were implemented in agreement with the PRISMA instructions [14]. The registration protocol was carried out in Prospero with the registration number CRD42021288077. The resulting PICOS framework performed is described in Table 1. The research question was: "Is the use of spectroscopic or camera method in dentistry accurate for color selection of the tooth?"

Study Selection
Two researchers (L.H. and R.B.) independently assessed the abstracts and titles of all the articles. Studies for full-text review were chosen based on the following eligibility criteria: (1) randomized controlled trial or in vitro papers studying the effect of using either digital shade selection or visual shade selection on the accuracy of color selection; (2) publishing in the English language; (3) assessing the accuracy of color measurement of the tooth and/or restorative material; (4) including a numerical value for the difference in color of the subject selected. Case series, reviews, case reports, and pilot studies were excluded. Full versions of any potentially suitable manuscripts were analyzed. Manuscripts that met the inclusion conditions or had inadequate data in the abstract and title to provide a clear judgment were considered for full evaluation. Full-text papers were evaluated by independent reviewers. Any variations in the decision-making process in regard to the suitability of the accepted articles was resolved and decided upon through the consensus of a third author (C.E.C.-S.). Only manuscripts that satisfied all of the eligibility criteria recorded were integrated for assessment.

Data Extraction
The information of concern from the papers included was organized using a harmonized form in a Microsoft Office Excel 2019 spreadsheet (Microsoft Corporation, Redmond, WA, USA). These records comprised demographic data (author and year of publication), method used for shade match, materials used, type of room light, the main outcome, and main results. The corresponding authors of the included articles were communicated twice via e-mail to regain any missing data. If the investigators did not answer within 10 days of the first communication, the omitted data was not incorporated.

Quality Assessment
The methodological quality of each involved manuscript was independently assessed by two investigators (R.B. and L.H.), considering the parameters of the Cochrane Collaboration's tool for evaluating risk of bias in randomized trials [15]. For the in vitro articles, the risk of bias in each manuscript was assessed according to the description of the subsequent parameters: specimen randomization, single operator, operator blinded, control group, standardized specimens, and sample size calculation [15]. If the reviewers stated the parameter, the study expected a "YES" of that precise parameter. In case of missing information, the parameter received a "NO." The risk of bias was categorized conferring to the sum of "YES" answers received: 1 to 2 designated a high bias, 3 to 4 medium, and 5 to 6 showed a low risk of bias. For the clinical studies, the risk of bias in each paper was considered according to the description of the subsequent parameters: allocation concealment, incomplete outcome data, sequence generation, blinding of the outcome assessors, selective outcome reporting, and other potential sources of bias. All these areas were assessed at the analysis level. Through risk of bias evaluation, any discrepancies between the researchers were solved through argument by accessing a third referee (C.E.C.-S.).

Statistical Analysis
Meta-analyses were completed using the Review Manager Software version 5.3.5 (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark). For the quantitative data, a random-effect model was conducted for evaluates, and pooledeffect estimates were acquired by comparing the standardized mean difference of the (∆E) between the classical shade guide and the computerized methods. Subgroup analyses were accomplished using a spectrophotometer or a digital camera to assess the shade selection. For the qualitative data, global analysis was conducted using a fixed-effects model, and pooled-effect estimates were acquired by comparing the risk difference of incorrect shade matching of the computerized methods (use of spectrophotometer or digital camera), or the conventional visual shade guide. A p-value of <0.05 was considered statistically significant. Statistical heterogeneity of the treatment effect among studies was evaluated using the Cochran Q test and the inconsistency I 2 test.

Results
A total of 4136 papers were obtained from the databases ( Figure 1). After removing the duplicates, the total quantity of literature found was 3299 publications for the primary examination. Then, 3268 manuscripts were excluded after revising the titles and summaries, leaving a total of 31 articles to be assessed for full-text review. Of these, 18 studies were excluded [16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33]. Exclusion reasons are assumed in the PRISMA flow diagram of the research, which resulted in a total of 13 manuscripts in the meta-analysis [34][35][36][37][38][39][40][41][42][43][44][45][46]. These manuscripts included seven in vitro studies and six clinical trials. The characteristics of the manuscripts incorporated in this systematic review were recapitulated in Table 3. Table 3. Demographic and study design data of the incorporated manuscripts.  The characteristics of the manuscripts incorporated in this systematic review were recapitulated in Table 3.    The meta-analysis of the quantitative data indicates that (∆E) between restorations where the shade matching was performed with the conventional method was greater than those where the shade matching was performed with the computerized methods (p = 0.007). According to the subgroup analysis, only the use of digital photographs for shade matching showed a reduction in (∆E) (p < 0.001), while the use of a spectrophotometer had no advantages over the use of visual shade guide tabs (p = 0.57) (Figure 2). On the other hand, for the qualitative analysis, global analysis proved that incorrect shade matching was higher when the conventional method using shade guide tabs was used (p < 0.001), irrespective of whether a spectrophotometer or a digital camera was used (p < 0.001) (Figure 3). On the other hand, for the qualitative analysis, global analysis proved that incorrect shade matching was higher when the conventional method using shade guide tabs was used (p < 0.001), irrespective of whether a spectrophotometer or a digital camera was used (p < 0.001) (Figure 3). On the other hand, for the qualitative analysis, global analysis proved that incorrect shade matching was higher when the conventional method using shade guide tabs was used (p < 0.001), irrespective of whether a spectrophotometer or a digital camera was used (p < 0.001) (Figure 3). Studying the methodological quality assessment parameters, most of the manuscripts involved were counted as having a medium risk of bias (Table 4 and Figure 4). However, several studies analyzed failed to account for the sample size calculation, single operator, and operator blinded parameters. Regarding the clinical studies, most of them were categorized as having a high risk of bias, since most of them failed to avoid performance and detection bias, reporting bias, bias due to incomplete data, and other bias. Studying the methodological quality assessment parameters, most of the manuscripts involved were counted as having a medium risk of bias (Table 4 and Figure 4). However, several studies analyzed failed to account for the sample size calculation, single operator, and operator blinded parameters. Regarding the clinical studies, most of them were categorized as having a high risk of bias, since most of them failed to avoid performance and detection bias, reporting bias, bias due to incomplete data, and other bias.

Discussion
This systematic review and meta-analysis were directed to compare different computerized methods with the visual shade guides for shade matching. The overall findings

Discussion
This systematic review and meta-analysis were directed to compare different computerized methods with the visual shade guides for shade matching. The overall findings revealed that the color differences of restorations performed using a visual shade guide were higher than with digital photography. In addition, the proportion of incorrect shade matching was higher when the visual shade guide tabs were used. Considering this, the null hypothesis stating that both digital and visual shade matching of the tooth were comparable was rejected.
It should be noted that shade matching in esthetic restorations was a key factor in the attainment of a perfect tooth color [47,48]. Furthermore, color mismatch in ceramic restorations can be very stressful for both clinicians and patients [49]. In this sense, the combination of visual and digital techniques has been suggested for a proper tooth color matching [35]. The visual technique consists of the use of shade guides to acquire the closest match to a natural tooth [5,7]. This way is subjective, and its precision is affected by several factors including experience, sex, and the need for training of the spectator, in addition to the tooth shade and light source [50][51][52]. Dental technicians should rely on a color match approach when operating only from an instruction based on a shade guide [53]. For this reason, the digital technique has been deemed more reliable than the visual technique alone [7]. Digital images achieved imperative data about tooth color crossways and the shape, surface, and characteristic features that the dental technician needs [35]. This could be explained by the improvement in communication and color selection [54]. Whenever feasible, both techniques should be combined [5,54].
In addition, mobile dental photography (MDP) seems to be favorable in terms of software applications, cost effectiveness, enhanced functionalities, and high-resolution photographs, which makes it suitable for acquiring color references [54,55]. The use of a smartphone with the help of the cross-polarizing filter of the smile Lite MDP allowed users to compare and evaluate the color matching amongst shade tabs [54,56,57].
One should bear in mind that by using this filter, the intrinsic shade variations of natural teeth for shade analysis might be revealed. In addition, it can help remove unwanted reflections and diffuse light created by the flash which can obscure features in the dental teeth, and hence produce issues in communication with the dental technician [18,21]. Consequently, referring to the results of Sampaio et al., [27] the use of cross-polarizing filters was the most standardized method for both communication and color assessment. It is important to mention that non-polarized tooth color images offered numerous glare designs on the surface of the tooth, which could be affected by the uneven distribution of saliva on tooth surfaces or roughness. These produce artifacts in tooth color image analysis. In juxtapose, a lack of glaring patterns was observed in the cross-polarized tooth color images [32].
It should be emphasized that (∆E) obtained by two measurements of the same object can be interpreted as acceptable or perceptible to the human eyes [58].
Acceptability means the agreement of the color of the restoration, while perceptibility indicates the detection of the (∆E) between both the tooth and the adjacent restoration [59]. A previous study suggested that a clinically acceptable level of visual perception was possible when (∆E = 3.7) in the CIELab system [58]. This system contains three coordinates: L* which refers to the luminosity of the object and fluctuates from black to white; a* which indicates the chroma in the green-red axis; and the b* which represents the chroma in the axis of blue-yellow [60]. Each color has a numerical value, delivering an objective description [27]. This might clarify the outcomes of this analysis obtained by a visual technique, as the preceding paper displayed that a non-acceptable restoration match was obtained with a higher (∆E) [35]. In this respect, clinicians should consider the digital technique when performing the shade-matching process. Digital images gave significant evidence about tooth color crossways, the surface, shape, and details that the dental laboratory technician needs [61,62]. Indeed, this could be possible when using a standardized light correcting device, thus helping to achieve a better result [59]. Both smartphones and digital cameras can promote a correct white balance by using a grey reference card with specific color coordinates to supply the color of the images in an accurate way [19,27,35]. This was considered successful in compensating the changes in tooth color triggered by various diffuser materials [19].
Additionally, a cross-polarizing filter might be assessed to eliminate specular reflections which can modify an image and lead to an imperfect analysis [27]. Hence, the use of less expensive instruments such as smartphones may enhance communication and help replicate the color in an easy, fast, and accurate way.
The percentage of dentists trusting the use of their smartphone cameras instead of professional DSLR cameras is rising exponentially because of the former's easy access and manipulation [63].
The findings of this analysis should be reflected with caution, since not all smartphones and digital cameras available on the market were tested. In addition, the evidence should be directed towards different light source devices which can facilitate communication between the dentist, patient, and the lab technician. Future research needs to consist of guided, principally randomized controlled clinical trials, with the aim of affording better vision into the performance of several computerized approaches to color-shade-matching success. Moreover, different spectrophotometers are already available in the market, and further studies testing these devices can provide a better understanding of color matching between the tooth and the restoration in an affordable way.

Conclusions
The use of digital photography and spectrophotometric measurements led to fewer color differences and less incorrect shade matching than conventional methods using color shade tabs.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ma15020468/s1, Table S1: Search strategy used in SCOPUS; Table  S2: Search strategy used in ISI Web of Science; Table S3: Search strategy used in the Cochrane Library; Table S4: Search strategy used in EMBASE.

Data Availability Statement:
The data that help the outcomes of this analysis are accessible from the corresponding authors upon reasonable request.