Abstract
Introduction. The purpose of this systematic review was to compare the accuracy of the three-dimensional images among different scanners, scanning techniques, and substrates. Materials and methods. Electronic databases (PubMed and Elsevier) were searched until March 2020. The systematic search was performed to identify the most precise method of obtaining a 3D image of the dentition. Results. Thirteen articles out of 221, considering the accuracy of 3D images, were selected. The main factors that are considered to have an influence on the precision are substrate type in the oral cavity, experience of the scanner’s operator, direct vs. indirect scanning, and the reproducibility of the procedure. Conclusion. Substrate type does have an impact on the overall accuracy of intraoral scans where dentin has the most and enamel the least accurately recorded dental structure. Experience of the operator has an influence on the accuracy, where more experienced operators and smaller scan sizes are made for more accurate scans. A conventional impression technique in a full-arch image provided the lowest deviation. The reproducibility of direct scanning was comparable to indirect scanning although a slight difference was noticeable (0.02 mm).
1. Introduction
Nowadays, 5 to 10% of dentists use the possibility to get a digital dental impression of the dentition with the use of intraoral scanners, and the percentage rises every year [1]. In the last decade, digitalization has gained increasing importance in the everyday dental practice [2]. Conventional impressions are still very common. However, it is well known that digital models offer more advantages, for example, better precision while measuring the size of teeth, calculating the orthodontic indices, and collecting all data needed for the diagnosis [3]. Due to the possibilities which digitally collected data provides, additional perspectives arise, which, compared to conventional methods, would be complicated or sometimes not even be possible [4]. Even though we are in the digital era of dentistry, a lot of clinicians doubt whether intraoral scan can fully replace an impression. A group of researchers from Switzerland concluded that the digital technique yielded higher local deviations while scanning the complete-arch cast [5]. A Korean group of prosthodontists claim that the accuracy of an intraoral scanner is greater than a conventional method. However, it decreases as the size of the scan increases [6]. A fundamental change and advantage for patients are digital impressions, especially for those with strong vomiting reflexes [7]. In orthodontics, proper diagnosis, treatment follow-up, and interdisciplinary consultations require often transport of plaster cast which causes potential damage. Additionally, the need of storage stone models, 10 years after finishing treatment, demands from the practitioners a lot of physical space in their dental offices [7]. Digital casts include more efficient storage and retrieval, easier transferability, superior durability, increased diagnostic versatility, and decreased processing time [8]. On the contrary, the advantages of conventional materials are that they are accurate, less expensive, and well-accepted [9, 10].
Therefore, the question is what kind of scanner and method of scanning should be used in digital orthodontics to get the best results?
2. Materials and Methods
This systematic review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The review was registered in PubMed and Elsevier databases. The literature search was independently conducted by two researchers (D.K., T. M) in March 2020, utilizing articles in English based on original studies. No time filter was used. The detailed search terms and strategy are presented in Table 1. The electronic search was complemented by a manual search of bibliographies from full-text articles and abstracts. Inclusion criteria were original articles in English, and exclusion criteria were systematic reviews, case reports, letters, and articles written in a different language than English.
2.1. PICOS Question
On the basis of Participants, Interventions, Comparisons, Outcomes, and Study (PICOS) design, the focus questions that guided this systematic review were which scanner is the most precise? What influences the precision of a digital image obtained from the scanning procedure? Does in vivo scanning provide better results than ex vivo scanning?
3. Results
Thirteen articles out of 221 considering the accuracy of 3D images were selected using inclusion criteria (Table 2). After reading all articles, only 13 of them were included in the study. Of the considered factors, the main ones that might have an influence on the precision are substrate type in the oral cavity, experience of the scanner’s operator, direct vs. indirect scanning, and the reproducibility of the procedure.
4. Discussion
A study group from Korea [13] compared two 3D images collected from in vivo and ex vivo scans made of casts based on impressions using the TRIOS scanner (3 Shape). The researchers compared the images of patients with a full dentition, except third molars, and scanned patients twice, with an interval of two weeks. The results showed greater discrepancies in the posterior than in the anterior region of dental arch on the scanned images. The average surface difference between the first and second images in the in vivo scans was about 0.02 mm greater than that of the ex vivo equivalents. The accuracy of measurements while comparing alginate impressions to dental scans is mostly based on recordings of the posterior and anterior regions (intercanine and intermolar distances). A study from Korea (2016) showed that there is no difference between plaster models and intraoral scans, except for one measurement of the lower intermolar width. The average surface difference amounts to 0.10 mm. Therefore, the results indicate that intraoral scans are acceptable clinically and can be used instead of plaster models [14]. A similar study from Brazil (2017) was undertaken, where not only intercanine and intermolar distances but also tooth diameter and height, overjet, overbite, and the sagittal relationship were measured. The findings showed that the measuring method can affect the reproducibility of the measurements [11]. The type of scanned tissues was always a challenge in terms of reliability when obtaining anatomical structures. In Nijmegen, in 2018, researchers studied whether the accuracy of the shape, color, and curvature of palatal soft tissues can be obtained in the scanned image. The results support the hypothesis that an intraoral scan can record a 3D image of palatal soft tissues [16].
4.1. Substrate
Another questionable factor that could affect the scan is the type of substrate. The type of scanned tissue (substrate) does not impact the overall accuracy of intraoral scans, which is a hypothesis that was confirmed by researchers from South Carolina in 2019 [15]. However, researchers from the same region, but a year later, claimed that the type of substrate does affect the trueness and precision of a scan. They found that active triangulation scanners are more sensitive to substrate differences than their parallel confocal counterparts. Due to the advances in technology, some scanners scan certain substrates better, but, in general, the new generation of scanners has overcome the problems of the old, including a collective improvement in the imaging of all substrates [17].
Digital orthodontics provides more comfort to the patient, which was also found in a study conducted in London in 2019. Patients questioned after the procedure of scanning and having their impressions taken answered that scanning is more comfortable than the latter. However, it takes more time, as the clinician has to calculate the automated PAR (Peer Assessment Rating) scoring [12].
There was a study that compared 2 intraoral scanners: TRIOS 3, 3 Shape and CS 3600, Carestream. TRIOS 3 displayed slightly higher precision (approximately 10 μm) compared to CS 3600, only after superimposition on the whole dental arch (). Both intraoral scanners showed good performance and comparable trueness (median: 0.0154 mm; ). However, in individual cases and in various, not spatially defined areas, higher imprecision was evident. Thus, the intraoral scanners’ appropriateness for highly demanding, spatially extended clinical applications remains questionable [18].
4.2. Experience of the Operator
Another aspect which is taken into account when comparing reproducibility and image trueness is the experience of the practitioner. Research from Seoul in 2018 showed that newer systems were less likely to be influenced by the length of clinical career as well as the region being scanned [19]. This theory was confirmed by studies from Seoul in 2017 and Brazil in 2020, though in the Brazilian study, the team of operators consisted of 3 professionals with different levels of experience in contrast to Seoul’s research studies, which used assistants as operators [20].
4.3. Type of Scanner or Software
A study group from Giessen studied the transfer accuracy of four different scanners (Trios3Cart, Trios3Pod, Trios4Pod, and Primescan), which were equipped with the latest software versions. They compared obtained data to conventional impressions. What they received was that current IOSs equipped with the latest software versions demonstrated less deviation for short-span distances compared to the conventional impression technique. However, the results showed that, for long-span distances, the conventional impression technique provided the lowest deviation. “Overall, currently available IOS systems demonstrated improvement regarding transfer accuracy of full-arch scans in patients” [21]. Another research that compared different scanners was carried out in Freiburg in 2013. They compared an intraoral scanner (iTero) and an extraoral scanner (D250; 3Shape, Copenhagen, Denmark) and found that iTero was less accurate than scanning with the D250. This result suggests that the intraoral conditions (saliva, limited spacing) contribute to the inaccuracy of a scan. Intraoral scanners could be used for treatment planning and manufacturing of tooth-supported appliances [22].
A study from New York compared 4 intraoral optical scanners (True Definition, TRIOS, CEREC Omnicam, and Emerald Scanner) on an edentulous mandible model with 6 hexagonal scan bodies. There were neither statistical nor clinical differences among scanners [23].
5. Conclusions
Substrate type has an impact on the overall accuracy of intraoral scans, where dentin is the most and enamel is the least accurately recorded dental structure. The experience of the operator has an influence on the accuracy, where more experienced operators and smaller scan sizes make more accurate scans. A conventional impression technique in a full-arch recording provides the lowest deviation. The reproducibility of direct scanning is comparable to indirect scanning although a slight difference can be noticed (0.02 mm).
Data Availability
No data were used to support the findings of the study.
Conflicts of Interest
The authors declare that they have no conflicts of interest.