Q U I N T E S S E N C E I N T E R N AT I O N A L
PROSTHODONTICS
Jae-Hong Kim
Quantitative evaluation of common errors in digital impression obtained by using an LED blue light in-office CAD/CAM system Jae-Hong Kim, PhD1/Ki-Baek Kim, PhD1/Sa-Hak Kim, PhD2/Woong-Chul Kim, PhD3/Hae-Young Kim, DDS, PhD4/ Ji-Hwan Kim, PhD3 Objective: This study addresses common errors that may occur during digital impression procedures using the CEREC AC and in-office CAD/CAM systems. Error types and frequencies resulting from digital impressions of the oral cavity were investigated and their origins identified to facilitate the acquisition of more accurate digital impressions. Method and Materials: A total of 1,251 digital impression cases, including 163 onlays and 1,088 inlays, were constructed as restorations using the CEREC AC system. Two evaluators determined five categories of digital errors as a basis for evaluation over two sessions. The five categories were as follows: inappropriate scanner positioning (ISP), improper handling of the scanner (IHS), irregular powder arrangement (IPA), improper cavity
preparation (ICP), and insufficient scanned data (ISD). Results: The most frequently encountered errors were non-linear powder application (IPA), inappropriate oral cavity scanner placement (ISP), and insufficient data (ISD). ISP showed that inlays had a slightly higher frequency of errors, but this observation was not statistically significant (P > .05). Conclusion: Most errors are caused by incorrect operation of intraoral scanners or uneven application of powder in acquiring digital impressions. To provide optimal digital impression results, careful chairside technique procedures and operation of intraoral scanners are required. (Quintessence Int 2015;46:401–407; doi: 10.3290/j.qi.a33685)
Key words: common error, digital impression, in-office CAD/CAM, intraoral scanner
The acquisition of an accurate impression is critical for successful restorative treatment. Despite its impor1
Research Professor, Department of Dental Laboratory Science and Engineering, College of Health Science, Korea University, Seoul, Republic of Korea.
2
Professor, Department of Dental Technology, School of Medical and Public Health, Kyungdong University, Gangwondo, Goseong, Republic of Korea.
3
Professor, Department of Dental Laboratory Science and Engineering, College of Health Science, Korea University, Seoul, Republic of Korea.
4
Associate Professor, Department of Dental Laboratory Science and Engineering, College of Health Science; and Department of Public Health Sciences, Graduate School and BK21+ Program in Public Health Sciences, Korea University, Seoul, Republic of Korea.
Correspondence: Professor Ji-Hwan Kim, Department of Dental Laboratory Science and Engineering, Korea University, 161 Jeongneungro, Seongbuk-gu, Seoul, 136-703 Republic of Korea. Email: kjh2804@ korea.ac.kr
VOLUME 46 • NUMBER 5 • MAY 2015
tance, in recent years there have not been any substantial improvements in the methods used by dentists to indirectly create dental restorations via traditional dental impressions. Shortcomings of this impression technique include imprecise capture of margins, the possibility of changes in impression material, volume changes in modeling materials, and contamination from saliva and blood, yet this method still remains the dominant practice.1 To rectify such quality deficiencies associated with manual production and to create consistent dental restorations, computer-aided design/ computer-aided manufacture (CAD/CAM) and automated technology have been introduced into the field
401
Q U I N T E S S E N C E I N T E R N AT I O N A L Kim et al
of dentistry.2,3 The advances in digital technology and incorporation of CAD/CAM systems have changed the processes of creating traditional hand-produced dental impressions and dental restorations.4 To address problems such as those mentioned above, the intraoral scanner was invented. This device reconstructs gingiva and tooth data using an oral scanner, allows for the observation of the constructed 3D model on a monitor, and occasionally designs and produces dental restorations in addition to sending the scanned data over the Internet to produce models via CAM technology. For the patient, there is the benefit of reducing the discomfort associated with a manual impression process, and since the dental engineering department receives precise patient data, the required time and material cost are decreased.5 In a single patient visit, an intraoral scanner can produce inlay, onlay, laminate, fixed partial denture (FPD) in monolithic ceramic, provisional restoration, and single crown by acquiring an impression digitally, replacing the conventional impression techniques and workflow. The connected milling machine used for the cutting process can immediately create the necessary dental restorations, which significantly decreases the time, cost, and patient discomfort.6 The main intraoral scanners that are widely used in dental clinical practice worldwide are as follows: CEREC system (Sirona Dental Systems), iTero (Cadent), Lava COS (3M ESPE), and E4D dentist (E4D Technologies). Intraoral scanners vary in operating principles. CEREC system scans are based on active triangulation measurements, iTero obtains images by parallel confocal analysis, Lava COS uses three-dimensional (3D)-inmotion technology, and E4D dentist uses a red laser technique.7 Intraoral scanners are recently developed optical impression devices that can generate a dental model directly from inside the mouth without the application of powder. These devices can also restructure the whole arch by overlapping individual dental models in real time. Although intraoral scanners, such as the CEREC system, iTero, Lava, and COS, are stated by the manufacturers to have a scanning performance of under 10 minutes when the soft tissue is involved and
402
the image data should be modified or an occlusal relationship should be determined, these scanners actually require more time than traditional devices for impression taking and plaster modeling. Trios (3Shape), which has recently been introduced, takes approximately 5 minutes, and the scanning is performed mainly along the occlusal surface of the maxilla and mandible. It can also restructure a more realistic dental model by integrating the data available on teeth and gum colors.8 Dental restorations are manufactured based on the acquired impressions; therefore, their accuracy is directly affected by the accuracy of the digital impressions. To date, few studies comparing esthetics and accuracy between digitally acquired impressions and traditionally acquired impressions have been reported, and most studies have determined the accuracy of digitally acquired impressions to be sufficient for clinical use.9-11 In 2010, Todorović et al12 evaluated 1,564 inlays, onlays, single crowns, and veneer dental restorations created by single-day treatments utilizing the CEREC system over a 5-year period and reported categories of error, including insufficient formation of the gingiva, instability of the oral scanner within the oral cavity, imprecise scanner positions and angles, and non-linear contrast spraying. However, there has not yet been a study quantitatively investigating the frequencies of errors associated with digital impressions. The aim of the present study is to evaluate digital impression data acquired from an in-office CAD/CAM system, quantitatively assess categories and frequencies of error, analyze sources of error, and ultimately, decrease errors and increase the accuracy associated with digital impressions.
METHOD AND MATERIALS This study was performed at the Seoul Dental Hospital from September 2012 to June 2013. The CEREC AC system was utilized to produce a total of 1,251 ceramic, inlay, and onlay dental restorations for single-day treatments, and the digital impression image data were analyzed. Software used to produce the dental restorations included CEREC 3D Version 3.65 (Sirona Dental
VOLUME 46 • NUMBER 5 • MAY 2015
Q U I N T E S S E N C E I N T E R N AT I O N A L Kim et al
Table 1
Interpretations for kappa values
Table 2
Distribution of the digital impression images (n, %) Molars
Premolars
Kappa statistic
Strength of agreement
Type of restoration
0
Poor
Onlay
44 (11.3)
63 (12.6)
31 (17.0)
25 (14.0)
163 (13.0)
0–0.2
Slight
Inlay
345 (88.7)
438 (87.4)
151 (83.0)
154 (86.0)
1,088 (87.0)
0.21–0.40
Fair
Total
389 (31.1)
501 (40.0)
182 (14.5)
179 (14.3)
1,251 (100.0)
0.41–0.60
Moderate
0.61–0.80
Substantial
0.81–1.00
Nearly perfect
Maxillary
Systems) and a CEREC AC with a Bluecam system (Sirona Dental Systems) used as a scanner. To investigate the sources of error, the most recently produced image data were observed three-dimensionally from the occlusal, mesial, distal, and gingival surfaces to categorize the errors. The error categories cited by Todorović et al12 were evaluated, adjusted, and divided into five types over two sessions by two evaluators with extensive experience in digital impressions. To prevent the first evaluator’s responses from influencing the second evaluator’s assessments, the second evaluator performed his evaluations 1 week after the first evaluator. The categories were as follows: • inappropriate scanner positioning (ISP) • improper handling of the scanner (IHS) • irregular powder arrangement (IPA) • improper cavity preparation (ICP) • insufficient scanned data (ISD). For example, when shades or dark parts are present on the inner wall of the cavity, and when the proximal contour, contact area, and cervical area of adjacent teeth are not clear, ISP was selected as an error category. In addition, when the overall digital impression resolution was low due to the unclear focus of the intraoral scanner (errors caused by flickering) and when an artifact was generated due to the damaged image near the cavity, the IHS category was selected. If there were duplicate errors within one image, they were counted twice in the individual categories. If there were two or more cavities within one digital image, then the cavity with the more complicated
VOLUME 46 • NUMBER 5 • MAY 2015
Mandibular
Maxillary
Mandibular
Total
structure and in a more posterior position was selected and analyzed. To categorize the digital impression data, frequencies and percentages were calculated using statistical analyses, and a chi-squared test was performed to assess the relationships between the errors in prosthesis positions and types. Cohen’s kappa values were calculated using the two sets of results by categorizing errors from the digital impression data to analyze inter-observer agreement (Table 1). Statistical significance was set at P < .05. The IBM SPSS Statistics version 20.0 software program (SPSS) was used for the statistical analysis.
RESULTS The digital impression image categories are included in Table 2. Of the posterior dental restorations considered, there were 163 onlays (13.0%) and 1,088 inlays (87.0%), 71% of which were located on molars. The calculated inter-observer agreement kappa value was 0.71, and this was statistically significant for both the maxilla and mandible (P < .05) (Table 3), thus suggesting consistency in the observations. According to the digital impression error analysis, of the total 1,251 images, IPA contributed to 264 (21.0%) errors, ISP to 212 (17.0%) errors, ISD to 211 (16.9%) errors, and ICP contributed to 80 (6.4%) errors, which represented the lowest proportion. Additionally, the distribution of relative frequencies varied according to the prosthesis position. Therefore, the relationship between prosthetic position and error was found to be statistically significant (P < .05).
403
Q U I N T E S S E N C E I N T E R N AT I O N A L Kim et al
Table 3
Inter-observer overall agreement (%) and kappa test results (κ) for assessing digital impression errors
Location
N
Overall agreement
Kappa coefficient (95% CI)
Maxilla
571
84.4
0.73 (0.59–0.90)
Mandible
680
75.0
0.69 (0.57–0.80)
Total
1,251
79.7
0.71 (0.58–0.84)
P value
.001
CI, confidence interval.
Table 4
Causes of digital impression errors according to location Reason for error*
Location
Any†
N
ISP
IHS
IPA
ICP
ISD
Premolar
182
94 (51.6%)
18 (9.9%)
2 (1.1%)
44 (24.2%)
31 (17.0%)
43 (23.6%)
Molar
389
325 (83.5%)
100 (25.7%)
75 (19.3%)
88 (22.6%)
25 (6.4%)
56 (14.4%)
Premolar
179
69 (38.5%)
25 (14.0%)
19 (10.6%)
19 (10.6%)
6 (3.4%)
31 (17.3%)
501
338 (67.5%)
69 (13.8%)
88 (17.6%)
113 (22.6%)
18 (3.6%)
81 (16.2%)
1,251
826 (66.0%)
212 (17.0%)
184 (14.7%)
264 (21.1%)
80 (6.4%)
211 (16.9%)
Maxillary
Mandibular Molar Total
*More than one reason could be selected. †The association between position and the presence of error(s) was significant for all five reasons (chi-squared test, P < .01).
Table 5
Proportion and statistical significance of digital impression errors according to restoration
Classification of restoration
No error
With error(s)*
Total
Onlay
62 (38.0%)
101 (62.0%)
163
Inlay
363 (33.4%)
725 (66.6%)
1088
Total
425 (34.0%)
826 (66.0%)
1251
P value†
.547
*For any error(s) among ISP, IHS, IPA, ICP, and ISD. †Obtained by chi-squared test.
The frequency of errors in the maxilla was higher than that in the mandible, and molars showed higher error frequencies than premolars (Table 4). Within the maxillary molars, of the 389 total errors, 25.7% resulted from ISP, and 19.3% resulted from IHS, displaying higher error frequencies than those in the maxillary premolars, mandibular premolars, and mandibular molars. For maxillary premolars, IPA and ISD contributed 24.2% and 23.6%, respectively, which were higher than the contributions from the maxillary molars or the mandibular portions. For the mandibular molars, IPA contributed the highest percentage (22.6%), followed by ISP (13.8%) for both the mandibular premolars and molars.
404
Overall, from a total of 1,251 images, 66.0% had at least 1 error, while 34.0% did not have any error. Frequencies of errors were higher for inlays (66.6%) versus onlays (62.0%), but the difference was not statistically significant (P > .05) (Table 5). For inlays and onlays, IPA errors were the most frequent, followed by ISP and ISD.
DISCUSSION This study investigated five categories of errors arising from digital impressions created from an in-office CAD/ CAM CEREC AC system used to complete dental restorations, which was subcategorized by the part and type of the prosthesis. The most frequently documented
VOLUME 46 • NUMBER 5 • MAY 2015
Q U I N T E S S E N C E I N T E R N AT I O N A L Kim et al
Distal
Mesial
Mesial
Buccal
Distal
Buccal
Lingual
Lingual
Fig 1 A scanned maxillary second molar with irregular powder arrangement (IPA).
Fig 2 A scanned maxillary first molar with inappropriate scanner positioning (ISP).
error was IPA, which contributed to 21.1% of the total data. Teeth are only partially reflective, which is why TiO2 powder is applied to prevent reflections from light rays. If more or less than the ideal amount of powder is applied, the precision of the digital impression is compromised, and it is therefore difficult to expect an accurate reconstruction. As shown in Fig 1, excessive powder application lowers the quality of the digital impression, and the surface of the adjacent teeth are roughly reconstructed without clarity and proper remodeling of the inside of the cavity. Powder application is a necessary component of the digital impression process and must be performed evenly despite salivary contamination, other obstacles such as the tongue, or patient discomfort. This requires both experience and skill from the technician.13 Consequently, CEREC Omnicam, the latest version, is powder-free, and it acquires oral images through video input in real-time and obtains digital impressions of the oral environment in real color.14 If a digital impression is obtained using iTero system or CEREC Omnicam, it has the advantage of scanning the natural state of the intraoral structure, because tooth coating (reflecting agent, such as titanium dioxide powder) is not required. Therefore, errors caused by uneven powder spray will no longer occur when the digital impression is taken. It may even be able to forestall negative effects on the marginal and internal fit of
dental restorations that are manufactured on the basis of an inaccurate digital impression caused by excessive use of powder.15 The second most common error was ISP, which represented 17.0% of the total error. According to the manufacturer, the scanner must be placed so that the teeth are present in the middle of the screen, at 3 mm to 4 mm distance, and at a 10-degree angle, without shadowing. The direction of the ray must be parallel to the long axis of the teeth when creating the impression, without touching the wall of the cavity. If there is too much of an angle, the ray will not reflect certain components of the cavity walls, which results in the partial loss of the impression.16,17 This principle also applies to adjacent teeth, and should be used to properly reconstruct the maximal boundary of the adjacent teeth. As shown in Fig 2, ISP resulted in the maxillary first molar being shadowed by adjacent teeth, thereby making it difficult to discern and resulting in a poor reconstruction. ISD represented 16.9% of the total results, which is similar to the frequency of ISP errors. For a precise and ideal design, sufficient information regarding both the abutment and adjacent teeth is required. However, in Fig 3 there is insufficient data for both regions, thus resulting in an incomplete reconstruction. IHS errors represented 14.7% of the total error. Such errors result from patient or technician fatigue or tech-
VOLUME 46 • NUMBER 5 • MAY 2015
405
Q U I N T E S S E N C E I N T E R N AT I O N A L Kim et al
Mesial
Distal
Buccal
Lingual
406
Buccal
Mesial
nician mistakes, which lead to unintentional tremor and subsequent blurring of the focus. The CEREC AC system has methods to overcome this error, including the incorporation of an ultraviolet light to adjust for tremor via an auto-focus mechanism.18 Figure 4 demonstrates a case in which IHS led to the capture of an out-of-focus digital impression, which resulted in the inability to reconstruct the overall structure of the teeth, including the marginal zones of the cavity.19 ICP represented 6.4% of the total error, which was comparatively low (Fig 5). Taking into account the straightforward direction of the beam, the cavity should be prepared to reflect the well-identified margins of the restorative material and cavities and form an 80- to 90-degree angle at the cavity margin for a precise impression.20 After an investigation including a sub-analysis of the various parts, errors of the molars were found to be more common than those of the premolars. It is possible that this could be explained by the fact that the scanner has less accessibility to the most posterior teeth, and therefore has difficulty obtaining their exact locations. Errors were more common in the maxilla than in the mandible, possibly due to the fact that the patient is lying down while the digital impression is obtained, which makes it more difficult for the scanner to match the maxilla. Digital impressions have time-saving advantages, but they also can decrease the chance of ill-fitting den-
Fig 4 A scanned mandibular dentition with improper handling of the scanner (IHS).
Distal
Fig 3 A scanned maxillary posterior tooth with insufficiency of scanned data (ISD).
Lingual Fig 5 An improper mesiodistal cavity preparation of the mandibular first molar (ICP).
tal restorations because the practitioner can immediately evaluate and eliminate preparation errors.21 However, ICP, IPA, and technician errors are very common in a clinical setting.22 Therefore, if these errors occur while obtaining digital impressions, they can be modified or supplemented via the “repair” menu in the CEREC system software. Moreover, it is considered desirable to take a new digital impression immediately after obtaining the patient’s consent by explaining the time-saving advantage of the digital impression process. This study categorized and expressed the frequencies of common errors made in digital impressions in the clinical environment, but represents images limited
VOLUME 46 • NUMBER 5 • MAY 2015
Q U I N T E S S E N C E I N T E R N AT I O N A L Kim et al
to the Seoul Dental Hospital and thus cannot be generalized. CEREC AC with a Bluecam system is being advertised as an improved system that can enable the general practitioner or assistant staff (dental hygienist) to acquire a digital impression with sufficient stability and accuracy after undergoing short training. As such, the process of obtaining digital impressions has innovatively been improved compared to the initial-stage CEREC system; however, for the parts requiring clinical treatment or experience for obtaining digital impressions, expert skills cannot be overlooked.23 Therefore, to ensure the accuracy of a digital impression, the influence of the learning curve should be considered. A general practitioner with little digital impression experience can face clinical issues with regards to the reproducibility that ensures coherent results or accuracy of digital impressions.24 Given that the data may vary according to the experience level of the general practitioner, there is a need to acquire and analyze additional digital impression data in a consistent fashion. This study is expected to provide useful information to general practitioners who want to use an intraoral scanner for patients in dental clinics. In this study, CEREC AC Bluecam was used. A follow-up study is warranted to evaluate the digital impressions obtained with CEREC Omnicam, which is capable of performing powder-free intraoral scanning.
CONCLUSION In the present study, which was performed under a limited set of conditions, the most frequent errors in acquiring digital impressions were the uneven application of powder, improper position of the intraoral scanner, and inadequate scan data. It is important to follow the recommendations in the treatment process because the accuracy of digital impressions using the CEREC system depends on practitioner technique and the characteristics of each patient. Because the precision of the digital impression directly affects the precision of the dental restoration, it is necessary to minimize the five categories of error investigated in this study to produce high-quality digital impressions.
VOLUME 46 • NUMBER 5 • MAY 2015
REFERENCES 1. Christensen GJ. The state of fixed prosthodontic impressions: room for improvement. J Am Dent Assoc 2005;136:343–346. 2. van Noort R. The future of dental devices is digital. Dent Mater 2012;28:3–12. 3. Miyazaki T, Hotta Y, Kunii J, Kuriyama S, Tamaki Y. A review of dental CAD/ CAM: current status and future perspectives from 20 years of experience. Dent Mater J 2009;28:44–56. 4. Christensen GJ. Impressions are changing: deciding on conventional, digital or digital plus in-office milling. J Am Dent Assoc 2009;140:1301–1304. 5. Birmbaum NS, Aaronson HB. Dental impressions using 3D digital scanners: virtual becomes reality. Comp Contin Educ Dent 2008;29:494–505. 6. Mehl C, Harder S, Byrne A, Kern M. Prosthodontics in digital times: a case report. Quintessence Int 2013;44:29–36. 7. Zandparsa R. Digital imaging and fabrication. Dent Clin North Am 2014;58:135–158. 8. Patzelt SB, Lamprinos C, Stampf S, Att W. The time efficiency of intraoral scanners: an in vitro comparative study. J Am Dent Assoc 2014;145:542–551. 9. Schaefer O, Decker M, Wittstock F, Kuepper H, Guentsch A. Impact of digital impression techniques on the adaption of ceramic partial crowns in vitro. J Dent 2014;42:677–683. 10. Tidehag P, Ottosson K, Sjögren G. Accuracy of ceramic restorations made using an in-office optical scanning technique: an in vitro study. Oper Dent 2014;39:308–316. 11. Seelbach P, Brueckel C, Wöstmann B. Accuracy of digital and conventional impression techniques and workflow. Clin Oral Investig 2013;17:1759–1764. 12. Todorović A, Lisjak D, Lazić V, Špadijer-Gostović A. Possible errors during the optical impression procedure. Serb Dent J 2010;57:30–37. 13. Fasbinder DJ. Clinical performance of chairside CAD/CAM restorations. J Am Dent Assoc 2006;137(Suppl 1):22S–31S. 14. Wiedhahn K, Schenk O, Fitzsche G. Cerec Omnicam-intraoral scan 2.0. Int J Comput Dent 2012;15:199–205. 15. Kim SY, Kim MJ, Han JS, Yeo IS, Lim YJ, Kwon HB. Accuracy of dies captured by an intraoral digital impression system using parallel confocal imaging. Int J Prosthodont 2013;26:161–163. 16. Wiedhahn K. The optical Cerec impression-electronic model production. Int J Comput Dent 1998;1:41–54. 17. Ender A, Mehl A. Influence of scanning strategies on the accuracy of digital intraoral scanning systems. Int J Comput Dent 2013;16:11–21. 18. Luthardt RG, Loos R, Quaas S. Accuracy of intraoral data acquisition in comparison to the conventional impression. Int J Comput Dent 2005;8:283–294. 19. Ender A, Mehl A. Accuracy of complete-arch dental impressions: a new method of measuring trueness and precision. J Prosthet Dent 2013;109:121– 128. 20. Mehl A, Ender A, Mörmann W, Attin T. Accuracy testing of a new intraoral 3D camera. Int J Comput Dent 2009;12:11–28. 21. Luthardt RG, Bornemann G, Lemelson S, Walter MH, Hüls A. An innovative method for evaluation of the 3-D internal fit of CAD/CAM crowns fabricated after direct optical versus indirect laser scan digitizing. Int J Prosthodont 2004;17:680–685. 22. Luthardt RG, Weber A, Rudolph H, Schone C, Quaas S, Walter M. Design and production of dental prosthetic restorations: basic research on dental CAD/ CAM technology. Int J Comput Dent 2002;5:165–176. 23. Batson ER, Cooper LF, Duqum I, Mendonça G. Clinical outcomes of three different crown systems with CAD/CAM technology. J Prosthet Dent 2014;112:770–777. 24. Giménez B, Ozcan M, Martinez-Rus F, Pradies G. Accuracy of a digital impression system based on active wavefront sampling technology for implants considering operator experience, implant angulation, and depth (Epub ahead of print 24 Jul 2013). Clin Implant Dent Relat Res, doi: 10.1111/cid.12124.
407
PROSTHODONTICS
Jae-Hong Kim
Quantitative evaluation of common errors in digital impression obtained by using an LED blue light in-office CAD/CAM system Jae-Hong Kim, PhD1/Ki-Baek Kim, PhD1/Sa-Hak Kim, PhD2/Woong-Chul Kim, PhD3/Hae-Young Kim, DDS, PhD4/ Ji-Hwan Kim, PhD3 Objective: This study addresses common errors that may occur during digital impression procedures using the CEREC AC and in-office CAD/CAM systems. Error types and frequencies resulting from digital impressions of the oral cavity were investigated and their origins identified to facilitate the acquisition of more accurate digital impressions. Method and Materials: A total of 1,251 digital impression cases, including 163 onlays and 1,088 inlays, were constructed as restorations using the CEREC AC system. Two evaluators determined five categories of digital errors as a basis for evaluation over two sessions. The five categories were as follows: inappropriate scanner positioning (ISP), improper handling of the scanner (IHS), irregular powder arrangement (IPA), improper cavity
preparation (ICP), and insufficient scanned data (ISD). Results: The most frequently encountered errors were non-linear powder application (IPA), inappropriate oral cavity scanner placement (ISP), and insufficient data (ISD). ISP showed that inlays had a slightly higher frequency of errors, but this observation was not statistically significant (P > .05). Conclusion: Most errors are caused by incorrect operation of intraoral scanners or uneven application of powder in acquiring digital impressions. To provide optimal digital impression results, careful chairside technique procedures and operation of intraoral scanners are required. (Quintessence Int 2015;46:401–407; doi: 10.3290/j.qi.a33685)
Key words: common error, digital impression, in-office CAD/CAM, intraoral scanner
The acquisition of an accurate impression is critical for successful restorative treatment. Despite its impor1
Research Professor, Department of Dental Laboratory Science and Engineering, College of Health Science, Korea University, Seoul, Republic of Korea.
2
Professor, Department of Dental Technology, School of Medical and Public Health, Kyungdong University, Gangwondo, Goseong, Republic of Korea.
3
Professor, Department of Dental Laboratory Science and Engineering, College of Health Science, Korea University, Seoul, Republic of Korea.
4
Associate Professor, Department of Dental Laboratory Science and Engineering, College of Health Science; and Department of Public Health Sciences, Graduate School and BK21+ Program in Public Health Sciences, Korea University, Seoul, Republic of Korea.
Correspondence: Professor Ji-Hwan Kim, Department of Dental Laboratory Science and Engineering, Korea University, 161 Jeongneungro, Seongbuk-gu, Seoul, 136-703 Republic of Korea. Email: kjh2804@ korea.ac.kr
VOLUME 46 • NUMBER 5 • MAY 2015
tance, in recent years there have not been any substantial improvements in the methods used by dentists to indirectly create dental restorations via traditional dental impressions. Shortcomings of this impression technique include imprecise capture of margins, the possibility of changes in impression material, volume changes in modeling materials, and contamination from saliva and blood, yet this method still remains the dominant practice.1 To rectify such quality deficiencies associated with manual production and to create consistent dental restorations, computer-aided design/ computer-aided manufacture (CAD/CAM) and automated technology have been introduced into the field
401
Q U I N T E S S E N C E I N T E R N AT I O N A L Kim et al
of dentistry.2,3 The advances in digital technology and incorporation of CAD/CAM systems have changed the processes of creating traditional hand-produced dental impressions and dental restorations.4 To address problems such as those mentioned above, the intraoral scanner was invented. This device reconstructs gingiva and tooth data using an oral scanner, allows for the observation of the constructed 3D model on a monitor, and occasionally designs and produces dental restorations in addition to sending the scanned data over the Internet to produce models via CAM technology. For the patient, there is the benefit of reducing the discomfort associated with a manual impression process, and since the dental engineering department receives precise patient data, the required time and material cost are decreased.5 In a single patient visit, an intraoral scanner can produce inlay, onlay, laminate, fixed partial denture (FPD) in monolithic ceramic, provisional restoration, and single crown by acquiring an impression digitally, replacing the conventional impression techniques and workflow. The connected milling machine used for the cutting process can immediately create the necessary dental restorations, which significantly decreases the time, cost, and patient discomfort.6 The main intraoral scanners that are widely used in dental clinical practice worldwide are as follows: CEREC system (Sirona Dental Systems), iTero (Cadent), Lava COS (3M ESPE), and E4D dentist (E4D Technologies). Intraoral scanners vary in operating principles. CEREC system scans are based on active triangulation measurements, iTero obtains images by parallel confocal analysis, Lava COS uses three-dimensional (3D)-inmotion technology, and E4D dentist uses a red laser technique.7 Intraoral scanners are recently developed optical impression devices that can generate a dental model directly from inside the mouth without the application of powder. These devices can also restructure the whole arch by overlapping individual dental models in real time. Although intraoral scanners, such as the CEREC system, iTero, Lava, and COS, are stated by the manufacturers to have a scanning performance of under 10 minutes when the soft tissue is involved and
402
the image data should be modified or an occlusal relationship should be determined, these scanners actually require more time than traditional devices for impression taking and plaster modeling. Trios (3Shape), which has recently been introduced, takes approximately 5 minutes, and the scanning is performed mainly along the occlusal surface of the maxilla and mandible. It can also restructure a more realistic dental model by integrating the data available on teeth and gum colors.8 Dental restorations are manufactured based on the acquired impressions; therefore, their accuracy is directly affected by the accuracy of the digital impressions. To date, few studies comparing esthetics and accuracy between digitally acquired impressions and traditionally acquired impressions have been reported, and most studies have determined the accuracy of digitally acquired impressions to be sufficient for clinical use.9-11 In 2010, Todorović et al12 evaluated 1,564 inlays, onlays, single crowns, and veneer dental restorations created by single-day treatments utilizing the CEREC system over a 5-year period and reported categories of error, including insufficient formation of the gingiva, instability of the oral scanner within the oral cavity, imprecise scanner positions and angles, and non-linear contrast spraying. However, there has not yet been a study quantitatively investigating the frequencies of errors associated with digital impressions. The aim of the present study is to evaluate digital impression data acquired from an in-office CAD/CAM system, quantitatively assess categories and frequencies of error, analyze sources of error, and ultimately, decrease errors and increase the accuracy associated with digital impressions.
METHOD AND MATERIALS This study was performed at the Seoul Dental Hospital from September 2012 to June 2013. The CEREC AC system was utilized to produce a total of 1,251 ceramic, inlay, and onlay dental restorations for single-day treatments, and the digital impression image data were analyzed. Software used to produce the dental restorations included CEREC 3D Version 3.65 (Sirona Dental
VOLUME 46 • NUMBER 5 • MAY 2015
Q U I N T E S S E N C E I N T E R N AT I O N A L Kim et al
Table 1
Interpretations for kappa values
Table 2
Distribution of the digital impression images (n, %) Molars
Premolars
Kappa statistic
Strength of agreement
Type of restoration
0
Poor
Onlay
44 (11.3)
63 (12.6)
31 (17.0)
25 (14.0)
163 (13.0)
0–0.2
Slight
Inlay
345 (88.7)
438 (87.4)
151 (83.0)
154 (86.0)
1,088 (87.0)
0.21–0.40
Fair
Total
389 (31.1)
501 (40.0)
182 (14.5)
179 (14.3)
1,251 (100.0)
0.41–0.60
Moderate
0.61–0.80
Substantial
0.81–1.00
Nearly perfect
Maxillary
Systems) and a CEREC AC with a Bluecam system (Sirona Dental Systems) used as a scanner. To investigate the sources of error, the most recently produced image data were observed three-dimensionally from the occlusal, mesial, distal, and gingival surfaces to categorize the errors. The error categories cited by Todorović et al12 were evaluated, adjusted, and divided into five types over two sessions by two evaluators with extensive experience in digital impressions. To prevent the first evaluator’s responses from influencing the second evaluator’s assessments, the second evaluator performed his evaluations 1 week after the first evaluator. The categories were as follows: • inappropriate scanner positioning (ISP) • improper handling of the scanner (IHS) • irregular powder arrangement (IPA) • improper cavity preparation (ICP) • insufficient scanned data (ISD). For example, when shades or dark parts are present on the inner wall of the cavity, and when the proximal contour, contact area, and cervical area of adjacent teeth are not clear, ISP was selected as an error category. In addition, when the overall digital impression resolution was low due to the unclear focus of the intraoral scanner (errors caused by flickering) and when an artifact was generated due to the damaged image near the cavity, the IHS category was selected. If there were duplicate errors within one image, they were counted twice in the individual categories. If there were two or more cavities within one digital image, then the cavity with the more complicated
VOLUME 46 • NUMBER 5 • MAY 2015
Mandibular
Maxillary
Mandibular
Total
structure and in a more posterior position was selected and analyzed. To categorize the digital impression data, frequencies and percentages were calculated using statistical analyses, and a chi-squared test was performed to assess the relationships between the errors in prosthesis positions and types. Cohen’s kappa values were calculated using the two sets of results by categorizing errors from the digital impression data to analyze inter-observer agreement (Table 1). Statistical significance was set at P < .05. The IBM SPSS Statistics version 20.0 software program (SPSS) was used for the statistical analysis.
RESULTS The digital impression image categories are included in Table 2. Of the posterior dental restorations considered, there were 163 onlays (13.0%) and 1,088 inlays (87.0%), 71% of which were located on molars. The calculated inter-observer agreement kappa value was 0.71, and this was statistically significant for both the maxilla and mandible (P < .05) (Table 3), thus suggesting consistency in the observations. According to the digital impression error analysis, of the total 1,251 images, IPA contributed to 264 (21.0%) errors, ISP to 212 (17.0%) errors, ISD to 211 (16.9%) errors, and ICP contributed to 80 (6.4%) errors, which represented the lowest proportion. Additionally, the distribution of relative frequencies varied according to the prosthesis position. Therefore, the relationship between prosthetic position and error was found to be statistically significant (P < .05).
403
Q U I N T E S S E N C E I N T E R N AT I O N A L Kim et al
Table 3
Inter-observer overall agreement (%) and kappa test results (κ) for assessing digital impression errors
Location
N
Overall agreement
Kappa coefficient (95% CI)
Maxilla
571
84.4
0.73 (0.59–0.90)
Mandible
680
75.0
0.69 (0.57–0.80)
Total
1,251
79.7
0.71 (0.58–0.84)
P value
.001
CI, confidence interval.
Table 4
Causes of digital impression errors according to location Reason for error*
Location
Any†
N
ISP
IHS
IPA
ICP
ISD
Premolar
182
94 (51.6%)
18 (9.9%)
2 (1.1%)
44 (24.2%)
31 (17.0%)
43 (23.6%)
Molar
389
325 (83.5%)
100 (25.7%)
75 (19.3%)
88 (22.6%)
25 (6.4%)
56 (14.4%)
Premolar
179
69 (38.5%)
25 (14.0%)
19 (10.6%)
19 (10.6%)
6 (3.4%)
31 (17.3%)
501
338 (67.5%)
69 (13.8%)
88 (17.6%)
113 (22.6%)
18 (3.6%)
81 (16.2%)
1,251
826 (66.0%)
212 (17.0%)
184 (14.7%)
264 (21.1%)
80 (6.4%)
211 (16.9%)
Maxillary
Mandibular Molar Total
*More than one reason could be selected. †The association between position and the presence of error(s) was significant for all five reasons (chi-squared test, P < .01).
Table 5
Proportion and statistical significance of digital impression errors according to restoration
Classification of restoration
No error
With error(s)*
Total
Onlay
62 (38.0%)
101 (62.0%)
163
Inlay
363 (33.4%)
725 (66.6%)
1088
Total
425 (34.0%)
826 (66.0%)
1251
P value†
.547
*For any error(s) among ISP, IHS, IPA, ICP, and ISD. †Obtained by chi-squared test.
The frequency of errors in the maxilla was higher than that in the mandible, and molars showed higher error frequencies than premolars (Table 4). Within the maxillary molars, of the 389 total errors, 25.7% resulted from ISP, and 19.3% resulted from IHS, displaying higher error frequencies than those in the maxillary premolars, mandibular premolars, and mandibular molars. For maxillary premolars, IPA and ISD contributed 24.2% and 23.6%, respectively, which were higher than the contributions from the maxillary molars or the mandibular portions. For the mandibular molars, IPA contributed the highest percentage (22.6%), followed by ISP (13.8%) for both the mandibular premolars and molars.
404
Overall, from a total of 1,251 images, 66.0% had at least 1 error, while 34.0% did not have any error. Frequencies of errors were higher for inlays (66.6%) versus onlays (62.0%), but the difference was not statistically significant (P > .05) (Table 5). For inlays and onlays, IPA errors were the most frequent, followed by ISP and ISD.
DISCUSSION This study investigated five categories of errors arising from digital impressions created from an in-office CAD/ CAM CEREC AC system used to complete dental restorations, which was subcategorized by the part and type of the prosthesis. The most frequently documented
VOLUME 46 • NUMBER 5 • MAY 2015
Q U I N T E S S E N C E I N T E R N AT I O N A L Kim et al
Distal
Mesial
Mesial
Buccal
Distal
Buccal
Lingual
Lingual
Fig 1 A scanned maxillary second molar with irregular powder arrangement (IPA).
Fig 2 A scanned maxillary first molar with inappropriate scanner positioning (ISP).
error was IPA, which contributed to 21.1% of the total data. Teeth are only partially reflective, which is why TiO2 powder is applied to prevent reflections from light rays. If more or less than the ideal amount of powder is applied, the precision of the digital impression is compromised, and it is therefore difficult to expect an accurate reconstruction. As shown in Fig 1, excessive powder application lowers the quality of the digital impression, and the surface of the adjacent teeth are roughly reconstructed without clarity and proper remodeling of the inside of the cavity. Powder application is a necessary component of the digital impression process and must be performed evenly despite salivary contamination, other obstacles such as the tongue, or patient discomfort. This requires both experience and skill from the technician.13 Consequently, CEREC Omnicam, the latest version, is powder-free, and it acquires oral images through video input in real-time and obtains digital impressions of the oral environment in real color.14 If a digital impression is obtained using iTero system or CEREC Omnicam, it has the advantage of scanning the natural state of the intraoral structure, because tooth coating (reflecting agent, such as titanium dioxide powder) is not required. Therefore, errors caused by uneven powder spray will no longer occur when the digital impression is taken. It may even be able to forestall negative effects on the marginal and internal fit of
dental restorations that are manufactured on the basis of an inaccurate digital impression caused by excessive use of powder.15 The second most common error was ISP, which represented 17.0% of the total error. According to the manufacturer, the scanner must be placed so that the teeth are present in the middle of the screen, at 3 mm to 4 mm distance, and at a 10-degree angle, without shadowing. The direction of the ray must be parallel to the long axis of the teeth when creating the impression, without touching the wall of the cavity. If there is too much of an angle, the ray will not reflect certain components of the cavity walls, which results in the partial loss of the impression.16,17 This principle also applies to adjacent teeth, and should be used to properly reconstruct the maximal boundary of the adjacent teeth. As shown in Fig 2, ISP resulted in the maxillary first molar being shadowed by adjacent teeth, thereby making it difficult to discern and resulting in a poor reconstruction. ISD represented 16.9% of the total results, which is similar to the frequency of ISP errors. For a precise and ideal design, sufficient information regarding both the abutment and adjacent teeth is required. However, in Fig 3 there is insufficient data for both regions, thus resulting in an incomplete reconstruction. IHS errors represented 14.7% of the total error. Such errors result from patient or technician fatigue or tech-
VOLUME 46 • NUMBER 5 • MAY 2015
405
Q U I N T E S S E N C E I N T E R N AT I O N A L Kim et al
Mesial
Distal
Buccal
Lingual
406
Buccal
Mesial
nician mistakes, which lead to unintentional tremor and subsequent blurring of the focus. The CEREC AC system has methods to overcome this error, including the incorporation of an ultraviolet light to adjust for tremor via an auto-focus mechanism.18 Figure 4 demonstrates a case in which IHS led to the capture of an out-of-focus digital impression, which resulted in the inability to reconstruct the overall structure of the teeth, including the marginal zones of the cavity.19 ICP represented 6.4% of the total error, which was comparatively low (Fig 5). Taking into account the straightforward direction of the beam, the cavity should be prepared to reflect the well-identified margins of the restorative material and cavities and form an 80- to 90-degree angle at the cavity margin for a precise impression.20 After an investigation including a sub-analysis of the various parts, errors of the molars were found to be more common than those of the premolars. It is possible that this could be explained by the fact that the scanner has less accessibility to the most posterior teeth, and therefore has difficulty obtaining their exact locations. Errors were more common in the maxilla than in the mandible, possibly due to the fact that the patient is lying down while the digital impression is obtained, which makes it more difficult for the scanner to match the maxilla. Digital impressions have time-saving advantages, but they also can decrease the chance of ill-fitting den-
Fig 4 A scanned mandibular dentition with improper handling of the scanner (IHS).
Distal
Fig 3 A scanned maxillary posterior tooth with insufficiency of scanned data (ISD).
Lingual Fig 5 An improper mesiodistal cavity preparation of the mandibular first molar (ICP).
tal restorations because the practitioner can immediately evaluate and eliminate preparation errors.21 However, ICP, IPA, and technician errors are very common in a clinical setting.22 Therefore, if these errors occur while obtaining digital impressions, they can be modified or supplemented via the “repair” menu in the CEREC system software. Moreover, it is considered desirable to take a new digital impression immediately after obtaining the patient’s consent by explaining the time-saving advantage of the digital impression process. This study categorized and expressed the frequencies of common errors made in digital impressions in the clinical environment, but represents images limited
VOLUME 46 • NUMBER 5 • MAY 2015
Q U I N T E S S E N C E I N T E R N AT I O N A L Kim et al
to the Seoul Dental Hospital and thus cannot be generalized. CEREC AC with a Bluecam system is being advertised as an improved system that can enable the general practitioner or assistant staff (dental hygienist) to acquire a digital impression with sufficient stability and accuracy after undergoing short training. As such, the process of obtaining digital impressions has innovatively been improved compared to the initial-stage CEREC system; however, for the parts requiring clinical treatment or experience for obtaining digital impressions, expert skills cannot be overlooked.23 Therefore, to ensure the accuracy of a digital impression, the influence of the learning curve should be considered. A general practitioner with little digital impression experience can face clinical issues with regards to the reproducibility that ensures coherent results or accuracy of digital impressions.24 Given that the data may vary according to the experience level of the general practitioner, there is a need to acquire and analyze additional digital impression data in a consistent fashion. This study is expected to provide useful information to general practitioners who want to use an intraoral scanner for patients in dental clinics. In this study, CEREC AC Bluecam was used. A follow-up study is warranted to evaluate the digital impressions obtained with CEREC Omnicam, which is capable of performing powder-free intraoral scanning.
CONCLUSION In the present study, which was performed under a limited set of conditions, the most frequent errors in acquiring digital impressions were the uneven application of powder, improper position of the intraoral scanner, and inadequate scan data. It is important to follow the recommendations in the treatment process because the accuracy of digital impressions using the CEREC system depends on practitioner technique and the characteristics of each patient. Because the precision of the digital impression directly affects the precision of the dental restoration, it is necessary to minimize the five categories of error investigated in this study to produce high-quality digital impressions.
VOLUME 46 • NUMBER 5 • MAY 2015
REFERENCES 1. Christensen GJ. The state of fixed prosthodontic impressions: room for improvement. J Am Dent Assoc 2005;136:343–346. 2. van Noort R. The future of dental devices is digital. Dent Mater 2012;28:3–12. 3. Miyazaki T, Hotta Y, Kunii J, Kuriyama S, Tamaki Y. A review of dental CAD/ CAM: current status and future perspectives from 20 years of experience. Dent Mater J 2009;28:44–56. 4. Christensen GJ. Impressions are changing: deciding on conventional, digital or digital plus in-office milling. J Am Dent Assoc 2009;140:1301–1304. 5. Birmbaum NS, Aaronson HB. Dental impressions using 3D digital scanners: virtual becomes reality. Comp Contin Educ Dent 2008;29:494–505. 6. Mehl C, Harder S, Byrne A, Kern M. Prosthodontics in digital times: a case report. Quintessence Int 2013;44:29–36. 7. Zandparsa R. Digital imaging and fabrication. Dent Clin North Am 2014;58:135–158. 8. Patzelt SB, Lamprinos C, Stampf S, Att W. The time efficiency of intraoral scanners: an in vitro comparative study. J Am Dent Assoc 2014;145:542–551. 9. Schaefer O, Decker M, Wittstock F, Kuepper H, Guentsch A. Impact of digital impression techniques on the adaption of ceramic partial crowns in vitro. J Dent 2014;42:677–683. 10. Tidehag P, Ottosson K, Sjögren G. Accuracy of ceramic restorations made using an in-office optical scanning technique: an in vitro study. Oper Dent 2014;39:308–316. 11. Seelbach P, Brueckel C, Wöstmann B. Accuracy of digital and conventional impression techniques and workflow. Clin Oral Investig 2013;17:1759–1764. 12. Todorović A, Lisjak D, Lazić V, Špadijer-Gostović A. Possible errors during the optical impression procedure. Serb Dent J 2010;57:30–37. 13. Fasbinder DJ. Clinical performance of chairside CAD/CAM restorations. J Am Dent Assoc 2006;137(Suppl 1):22S–31S. 14. Wiedhahn K, Schenk O, Fitzsche G. Cerec Omnicam-intraoral scan 2.0. Int J Comput Dent 2012;15:199–205. 15. Kim SY, Kim MJ, Han JS, Yeo IS, Lim YJ, Kwon HB. Accuracy of dies captured by an intraoral digital impression system using parallel confocal imaging. Int J Prosthodont 2013;26:161–163. 16. Wiedhahn K. The optical Cerec impression-electronic model production. Int J Comput Dent 1998;1:41–54. 17. Ender A, Mehl A. Influence of scanning strategies on the accuracy of digital intraoral scanning systems. Int J Comput Dent 2013;16:11–21. 18. Luthardt RG, Loos R, Quaas S. Accuracy of intraoral data acquisition in comparison to the conventional impression. Int J Comput Dent 2005;8:283–294. 19. Ender A, Mehl A. Accuracy of complete-arch dental impressions: a new method of measuring trueness and precision. J Prosthet Dent 2013;109:121– 128. 20. Mehl A, Ender A, Mörmann W, Attin T. Accuracy testing of a new intraoral 3D camera. Int J Comput Dent 2009;12:11–28. 21. Luthardt RG, Bornemann G, Lemelson S, Walter MH, Hüls A. An innovative method for evaluation of the 3-D internal fit of CAD/CAM crowns fabricated after direct optical versus indirect laser scan digitizing. Int J Prosthodont 2004;17:680–685. 22. Luthardt RG, Weber A, Rudolph H, Schone C, Quaas S, Walter M. Design and production of dental prosthetic restorations: basic research on dental CAD/ CAM technology. Int J Comput Dent 2002;5:165–176. 23. Batson ER, Cooper LF, Duqum I, Mendonça G. Clinical outcomes of three different crown systems with CAD/CAM technology. J Prosthet Dent 2014;112:770–777. 24. Giménez B, Ozcan M, Martinez-Rus F, Pradies G. Accuracy of a digital impression system based on active wavefront sampling technology for implants considering operator experience, implant angulation, and depth (Epub ahead of print 24 Jul 2013). Clin Implant Dent Relat Res, doi: 10.1111/cid.12124.
407
