- Email: [email protected]
In vitro model to evaluate reliability and accuracy of a dental shade-matching instrument
In vitro model to evaluate reliability and accuracy of a dental shade-matching instrument Seungyee Kim-Pusateri, DDS, MS,a Jane D. Brewer, DDS, MS,b Robert G. Dunford, MA,c and Alvin G. Wee, BDS, MS, MPHd State University of New York at Buffalo, Buffalo, NY; Nebraska Medical Center, Omaha, Neb Statement of problem. There are several electronic shade-matching instruments available for clinical use; unfortunately, there are limited acceptable in vitro models to evaluate their reliability and accuracy. Purpose. The purpose of this in vitro study was to evaluate the reliability and accuracy of a dental clinical shadematching instrument. Material and methods. Using the shade-matching instrument (ShadeScan), color measurements were made of 3 commercial shade guides (VITA Classical, VITA 3D-Master, and Chromascop). Shade tabs were selected and placed in the middle of a gingival matrix (Shofu Gummy), with tabs of the same nominal shade from additional shade guides placed on both sides. Measurements were made of the central region of the shade tab inside a black box. For the reliability assessment, each shade tab from each of the 3 shade guide types was measured 10 times. For the accuracy assessment, each shade tab from 10 guides of each of the 3 types evaluated was measured once. Reliability, accuracy, and 95% confidence intervals were calculated for each shade tab. Differences were determined by 1-way ANOVA followed by the Bonferroni multiple comparison procedure. Results. Reliability of ShadeScan was as follows: VITA Classical = 95.0%, VITA 3D-Master = 91.2%, and Chromascop = 76.5%. Accuracy of ShadeScan was as follows: VITA Classical = 65.0%, VITA 3D-Master = 54.2%, Chromascop = 84.5%. Conclusions. This in vitro study showed a varying degree of reliability and accuracy for ShadeScan, depending on the type of shade guide system used. (J Prosthet Dent 2007;98:353-358)
Clinical Implications
Visual shade confirmation is still recommended when using the clinical dental shade-matching device tested, due to anatomical variations of natural teeth that may potentially influence reliability and accuracy.
Presented in part at the International Association of Dental Research Meeting, Honolulu, Hawaii, March, 2004, and the Annual Carl O. Boucher Conference, Columbus, Ohio, April, 2004. Supported in part by a grant from the Editorial Council for The Journal of Prosthetic Dentistry and by an equipment loan from Cynovad, Saint-Laurent, Canada. Assistant Professor, Department of Restorative Dentistry, State University of New York at Buffalo. Clinical Associate Professor, Department of Restorative Dentistry, State University of New York at Buffalo. c Scientific Programmer, Department of Oral Biology, State University of New York at Buffalo. d Associate Professor and Director, Division of Oral Facial Prosthetics/Dental Oncology, Department of Otolaryngology, Nebraska Medical Center. a
b
Kim-Pusateri et al
354
Volume 98 Issue 5 There are several problems with making clinical shade selection. Shade selection for dental restorations is affected by many factors, resulting in frequent errors in appearance-matching of restorations to natural teeth. When shades are selected by a human observer, the color-matching process is subjective.1,2 Natural teeth have several characteristics that make shade selection difficult. For example, natural teeth have a double-layering effect caused by enamel translucency and dentin opacity.3 Factors affecting human perception of color, such as lighting conditions and human physiologic variables, cause limitations in traditional visual shade selection.4 The increased emphasis on dental esthetics in recent years has driven the search for a more accurate shade-replication procedure than the present methods provide. Currently available electronic shade-matching devices are spectrophotometers, colorimeters, digital color analyzers, or combinations of these. Spectrophotometers are useful in the measurement of surface color. A prism disperses white light from a tungsten-filament bulb in the spectrophotometer into a spectrum of wavelength bands between 10 and 20 nm.5 The amount of light reflected from a specimen is measured for each wavelength in the visible spectrum. Of all devices, a spectrophotometer is the most accurate for absolute color measurement. These instruments have a longer working life than colorimeters and are unaffected by object metamerism.5,6 Colorimeters are useful in quantifying color differences between specimens. These devices measure tristimulus values according to CIE illuminant and observer conditions.7 Colorimeters use photodiode filters to control light reaching the specimen. The light reflected from the specimen is then measured by a detector. Colorimeters are easier to use and are less expensive than spectrophotometers. However, repeatability may be poor due to aging of filters, and object metamerism can be a chal-
1 ShadeScan (Cynovad) device. lenge to their accuracy.6 ShadeScan (Cynovad, Saint-Laurent, Canada) (Fig. 1) combines a colorimeter with digital imaging. It consists of a handpiece with a liquid crystal display screen which can be seated in a base unit, allowing autocalibration. It contains a halogen light source and uses 45/0 measuring geometry. The device can be used under any lighting condition because the halogen light is bright enough to be unaffected by ambient light.6 The downloaded data identifies the shade closest to the measured specimen. The database of the ShadeScan (Cynovad) software system supports the following shade guides: VITA Classical (VITA Zahnfabrik, Bad Sackingen, Germany), VITA 3D-Master (VITA Zahnfabrik), Ivoclar Chromascop (Ivoclar Vivadent, Schaan, Liechtenstein), Shofu Vintage Halo (Shofu Dental, San Marcos, Calif ), Noritake (Noritake Dental Supply Co Ltd, Nagoya, Japan), and Dentsply EsthetX (Dentsply Caulk, Milford, Del). Currently, there are several electronic shade-matching instruments available for clinical use,8,9,10 but there are limited acceptable in vitro models for evaluating their reliability and accuracy. As more shade-measuring de-
The Journal of Prosthetic Dentistry
vices become commercially available, a standard in vitro model would be useful for predicting their success in clinical practice. The purpose of this study was to evaluate the reliability and accuracy of a newly introduced clinical shade-matching instrument using an in vitro model developed for this investigation. The null hypothesis tested was that there is no difference in reliability and accuracy of the shade-matching instrument when measuring 3 different commercial shade guides.
MATERIAL AND METHODS This in vitro evaluation of a shadematching instrument was based on measurements of shade tabs in a simulated oral environment. Using the ShadeScan (Cynovad) instrument, color measurements were made of 3 commercial shade guides: (1) VITA Classical (VITA Zahnfabrik), (2) VITA 3-D Master (VITA Zahnfabrik), and (3) Chromascop (Ivoclar Vivadent). Ten shade guides from each manufacturer were obtained though convenience sampling from 2 institutions. All shade tabs were first cleaned using an ultrasonic cleaning solution (General Purpose Cleaner, nonammoni-
Kim-Pusateri et al
355
November 2007 ated; L&R Mfg Co, Kearny, NJ) for 15 minutes. They were later cleaned using an ultrasonic cleaner for another 15 minutes. Prior to color measurements, shade tabs were cleaned with soap (Ultra Ivory; Proctor & Gamble, Cincinnati, Ohio) and, subsequently, with distilled water. Simulated clinical conditions were created for color measurement of each shade tab. The shade tab to be measured was placed in the middle of a medium gingiva-colored matrix (Shofu Gummy; Shofu Dental Corp) with identically colored shade tabs placed on either side to simulate adjacent teeth (Fig. 2). The shade measurements were made within a black matte box (19.0 cm x 12.7 cm x 9.5 cm) to simulate the oral cavity. A positioning jig was used to accurately reposition the measuring unit for each shade measurement. The shade-matching device was operated according to the manufacturer’s instructions. For the reliability aspect of the study, each shade tab from 1 VITA Classical (VITA Zahnfabrik) shade guide (16 shade tabs), 1 VITA 3DMaster (VITA Zahnfabrik) shade guide (26 shade tabs), and 1 Chromascop (Ivoclar Vivadent) shade guide (20 shade tabs) were measured 10 nonconsecutive times. Identical repeated shade measurements were considered to be reliable whether or not the instrument measurement matched the actual shade tab. The mean reliability for each shade guide and the overall mean reliability of the shade-matching device were calculated. For the accuracy aspect of the study, each shade tab from 30 shade guides (10 VITA Classical; VITA Zahnfabrik, 10 VITA 3D-Master; VITA Zahnfabrik, and 10 Chromascop; Ivoclar Vivadent) were measured once. Accurate measurements were defined as identical shade matches to the shade tab. The accuracy of measurements was based on whether the shade-matching device selected a shade identical to the shade tab measured. The mean accuracy for each
Kim-Pusateri et al
2 Experimental apparatus. shade guide and the overall mean accuracy of the shade-matching device were calculated. Images were downloaded and analyzed with ShadeScan software (ShadeScan Software Version 3.1.0; Cynovad). For reliability, each shade tab was assessed 10 times for each of the 3 shade guides. The percent agreement of each tab was calculated and the 3 guides were compared by 1-way analysis of variance (ANOVA) followed by the Bonferroni multiple comparison procedure.11 Accuracy was assessed by using 10 guides for each of the 3 types of shade guides. The percent correct was calculated for each and analyzed by a 1-way ANOVA followed by the Bonferroni procedure. Ninetyfive percent confidence intervals for mean reliability and mean accuracy were calculated. Overall, a 5% level of significance was used for ANOVA. Multiple comparisons were not significant unless the corresponding P value was less than .0167. All statistical analyses and calculations were completed with statistical software (StatView statistical program version 5.0.1; SAS Institute, Inc, Cary, NC).
RESULTS
A total of 620 color measurements were made for the reliability evaluation (62 shade tabs measured 10 times) and 620 color measurements were made for the accuracy evaluation (620 shade tabs measured
once). The mean reliability for each shade guide is shown in Table I. Measurements of VITA Classical (VITA Zahnfabrik) produced the highest average percent reliability (95.0%), with 3D-Master (VITA Zahnfabrik) slightly (but not significantly) less at 91.2%. Measurements of Chromascop (Ivoclar Vivadent) were significantly less reliable at 76.5%. The analysis of reliability (Table II) revealed significant differences among the 3 shade guides (ANOVA, P=.002). The Fisher’s PLSD (Table III) indicates that 2 guides were significantly different from one other. The result of this multiple comparison procedure is that the mean reliability of measurements of Chromascop (Ivoclar Vivadent) was significantly lower than the other shade guides evaluated. The mean accuracy for each shade guide is shown in Table IV. Measurements of 3D-Master (VITA Zahnfabrik) produced the lowest average percent of accuracy (54.2%), with VITA Classical (VITA Zahnfabrik) slightly higher at 65.0%. Measurements of Chromascop (Ivoclar Vivadent) produced the highest accuracy (84.5%). The analysis of accuracy (Table V) revealed significant differences among the 3 shade guides (ANOVA, P<.001). The Fisher’s PLSD (Table VI) indicates that all 3 guides were significantly different from each other. This multiple comparison procedure demonstrated that the mean accuracy measurements of all 3 shade guides was significantly different.
356
Volume 98 Issue 5
Table I. Mean (SD) for reliability Shade Guide
Count
Mean (SD)
95% Confidence Interval
VITA Classical
16
95.0 (10.95)
(87.65, 100.00*)
3D-Master
26
91.2 (16.33)
(82.56, 99.75)
Chromascop
20
76.5 (19.81)
(64.61, 88.39)
* Actual estimated upper bound is truncated at 100%
Table II. ANOVA for reliability df
Sum of Squares
Mean Square
F
P
Shade guide
2
3666.7
1833.4
6.794
.002
Residual
59
15920.4
269.8
Table III. Fisher’s PLSD for reliability (significance level, 5%) Mean Difference
Critical Difference
P
VITA Classical vs. 3D-Master
3.846
12.862
.464
VITA Classical vs. Chromascop
18.500
13.577
.001
3D-Master vs. Chromascop
14.654
12.039
.004
Shade Guides
Table IV. Mean (SD) for accuracy (n=10) Shade Guide
Mean (SD)
95% Confidence Interval
VITA Classical
65.0 (7.91)
(58.29, 71.71)
3D-Master
54.2 (4.23)
(50.64, 57.83)
Chromascop
84.5 (13.63)
(72.93, 96.07)
Table V. ANOVA for accuracy df
Sum of Squares
Mean Square
F
P
Shade guide
2
4708.2
2354.1
26.525
<.001
Residual
27
2396.2
88.8
The Journal of Prosthetic Dentistry
Kim-Pusateri et al
357
November 2007
Table VI. Fisher’s PLSD for accuracy (significance level, 5%) Mean Difference
Critical Difference
P
VITA Classical vs. 3D-Master
10.769
10.754
.017
VITA Classical vs. Chromascop
–19.500
10.754
<.001
3D-Master vs. Chromascop
–30.269
10.754
<.001
Shade Guides
DISCUSSION The null hypothesis that there are no differences in reliability among the 3 types of shade guides measured was rejected (ANOVA, P=.002) (Table II). The mean reliability of measurements of Chromascop (Ivoclar Vivadent) shade guides was significantly lower than the other shade guides. The null hypothesis that there are no differences in accuracy among the 3 types of shade guides measured was also rejected (ANOVA, P<.001) (Table V). The mean accuracy measurements of all 3 shade guides were significantly different (P<.001). An important concern of electronic shade-matching devices is their ability to reliably and accurately measure tooth color. Reliability refers to the consistency of the output of the device when measuring a given standard. Accuracy refers to the ability of the device to provide a correct color match for a sample. A standard protocol to evaluate the reliability and accuracy of emerging shade-measuring devices would be beneficial. Since there is no established standard to compare measurements, it has been proposed that shade guides serve as the standard for color measurement.12 In the present study, an in vitro model was developed using shade tabs in a simulated oral environment for the evaluation of electronic shade-matching devices. The use of shade tabs for the target sample offers several advantages. Shade tabs are representative of natural teeth in terms of size, contour, and color. Shade tabs are also readily available and reproducible, assuming an insignificant variation in shade guides from the same manufacturer.
Kim-Pusateri et al
Other substrates with more clinical relevance could have been selected for measurement, such as extracted teeth or ceramic restorations. For the purpose of this study, it was assumed that all shade tabs that were selected represented the variation of the shade guides available from the manufacturers. Shade tabs may vary among shade guides from the same manufacturer, and this was a consideration in measuring the 10 sets of selected shade guides. Inherent differences among the shade tabs may be reflected as incorrect color measurements. Additionally, the irregular surface of a shade tab may affect the spectral reflectance of the shade-matching device, and consequently, the output.13 These potential variations were not evaluated in this study. In this evaluation, reliability and accuracy were the 2 objective measures evaluated. Reliability was defined as the percentage of identical repeated shade measurements, whether or not the measurement matched the actual shade tab. There was no significant difference between VITA Classical (VITA Zahnfabrik) and VITA 3D-Master (VITA Zahnfabrik) shade guides in terms of reliability (P=.4641). However, the reliability between VITA Classical and Chromascop (Ivoclar Vivadent) (P=.001), and VITA 3D-Master and Chromascop (P=.004), was statistically different. Chromascop shade tabs are smaller than both VITA 3D-Master and VITA Classical shade tabs. When Chromascop shade tabs were placed in the Shofu Gummy, spaces existed on either side of the target shade tab which were not present with VITA 3D-Master or VITA Classical shade tabs. This may have
influenced the reliability of ShadeScan (Cynovad) in the measurement of Chromascop shade tabs and is a limitation of the study. For the accuracy evaluation of ShadeScan (Cynovad), accurate measurements were identified as identical shade matches to the shade tab. There were significant differences among the 3 shade guides in terms of accuracy (Table IV). The wide range in accuracy (54.2% to 84.5%) illustrates that variation among shade tabs from different manufacturers influences the accuracy of shade measurements. Variation among natural teeth would be expected in a clinical setting, lending support for the use of multiple shade guides when performing an in vitro evaluation of electronic shadematching devices. The ease of use and degree of technique sensitivity were not evaluated in this study. These factors are not as important under the controlled conditions of this study, but may be important under clinical conditions. For example, holding this device steady is more difficult intraorally compared to use of a positioning jig. Movements made during shade measurement may impair accurate shade measurement. When tooth shade is measured by an electronic shade-matching device, color is expressed in terms of standards established by manufacturers of dental shade guides. The shade tab is therefore the communication tool between dentists and dental laboratory technicians regardless of whether tooth shade is measured visually or with an electronic device. Successful fabrication of a restoration requires the technician to match the color of the definitive ceramic restoration to
358
Volume 98 Issue 5 the prescribed shade tab. However, an imperceptible variation in color matching between a natural tooth and shade tab may become significant if the restoration varies further from the shade tab. Shade tabs serve as an intermediate step and therefore introduce 2 possibly conflicting sources of error – by the dentist in shade selection and by the technician in shade replication. This study does not allow for conclusions to be made regarding the clinical performance of the device tested, since no measurements were made on natural teeth. There are several factors that may influence accuracy and reliability of this device clinically. Unlike the shade tabs used in this study, natural teeth have variable surface texture and anatomic variations that may influence shade measurement. As discussed previously, natural teeth have a 3-dimensional double-layering effect and varying degrees of translucency compared to dental guides.3 Additional in vivo studies are needed before a prediction of clinical performance can be made. Only 1 instrument was evaluated in this study. To determine the overall usefulness of this data, other instruments are being evaluated in an ongoing study following this protocol.
CONCLUSIONS Within the limitations of this study, the following conclusions were made:
1. There was no significant difference in reliability measurements of VITA Classical (95.0%) and VITA 3D-Master (91.2%) shade guides (P=.4641). 2. Significant differences were found with reliability measurements of VITA Classical (95.0%) and Chromascop (76.5%) shade guides (P=.001), as well as VITA 3D-Master (91.2%) and Chromascop (76.5%) shade guides (P=.004). 3. The mean reliability of measurements of Chromascop (Ivoclar Vivadent) shade guides was significantly lower than the other shade guides. 4. Accuracy measurements of VITA 3D-Master produced the lowest average percent accuracy (54.2%), with VITA Classical slightly higher at 65.0%. Measurements of Chromascop produced the highest accuracy (84.5%). 5. The mean accuracy measurements of all 3 shade guides were significantly different (P<.001).
REFERENCES 1. Culpepper WD. A comparative study of shade-matching procedures. J Prosthet Dent 1970;24:166-73. 2. Geary JL, Kinirons MJ. Colour perception of laboratory-fired samples of body-coloured ceramic. J Dent 1999;27:145-8. 3. O’Brien WJ. Double layer effect and other optical phenomena related to esthetics. Dent Clin North Am 1985;29:667-72. 4. Carsten DL. Successful shade matching – what does it take? Compend Contin Educ Dent 2003;24:175-8, 180, 182. 5. Berns RS. Billmeyer and Saltzman’s principles of color technology. 3rd ed. New York: John Wiley & Sons Inc; 2000. p.
88-92. 6. Paravina RD, Powers JM. Esthetic color training in dentistry. St. Louis: Mosby; 2004. p. 17-28, 169-170. 7. CIE (Commission Internationale de l’Eclairage). Colorimetry, official recommendations of the International Commission on Illumination, Publication CIE No.15:2004: Colorimetry, 3rd ed. 8. Paul S, Peter A, Pietrobon N, Hammerle CH. Visual and spectrophotometric shade analysis of human teeth. J Dent Res 2002;81:578-82. 9. Yap AUJ, Sim CPC, Loh WL, Teo JH. Human-eye versus computerized color matching. Oper Dent 1999;24:358-63. 10.Okubo SR, Kanawati A, Richards MW, Childress S. Evaluation of visual and instrument shade matching. J Prosthet Dent 1998;80:642-8. 11.Fleiss JL. The design and analysis of clinical experiments. New York: John Wiley & Sons Inc; 1986. p. 8-13, 46-88. 12.Dozic A, Kleverlaan CJ, El-Zohairy A, Feilzer AJ, Khashayar G. Performance of five commercially available tooth color-measuring devices. J Prosthodont 2007;16:93-100. 13.Seghi RR. Effects of instrument-measuring geometry on colorimetric assessments of dental porcelains. J Dent Res 1990;69:1180-3. Corresponding author: Dr Seungyee Kim-Pusateri State University of New York at Buffalo Department of Restorative Dentistry School of Dental Medicine 3435 Main St. Squire Hall Room 222 Buffalo, NY 14214 Fax: 716-829-2440 E-mail: [email protected] Acknowledgements The authors thank Cynovad (Saint-Laurent, Canada), Ms April Logue, former Research Associate at Section of Restorative and Prosthetic Dentistry, for assistance, and Corey E. Valentine for assistance with ShadeScan measurement. Copyright © 2007 by the Editorial Council for The Journal of Prosthetic Dentistry.
Availability of Journal Back Issues As a service to our subscribers, copies of back issues of The Journal of Prosthetic Dentistry for the preceding 5 years are maintained and are available for purchase from Elsevier Inc until inventory is depleted. Please write to Elsevier Inc, Subscription Customer Service, 6277 Sea Harbor Dr, Orlando, FL 32887, or call 800-654-2452 or 407-345-4000 for information on availability of particular issues and prices.
The Journal of Prosthetic Dentistry
Kim-Pusateri et al
Clinical Implications
Visual shade confirmation is still recommended when using the clinical dental shade-matching device tested, due to anatomical variations of natural teeth that may potentially influence reliability and accuracy.
Presented in part at the International Association of Dental Research Meeting, Honolulu, Hawaii, March, 2004, and the Annual Carl O. Boucher Conference, Columbus, Ohio, April, 2004. Supported in part by a grant from the Editorial Council for The Journal of Prosthetic Dentistry and by an equipment loan from Cynovad, Saint-Laurent, Canada. Assistant Professor, Department of Restorative Dentistry, State University of New York at Buffalo. Clinical Associate Professor, Department of Restorative Dentistry, State University of New York at Buffalo. c Scientific Programmer, Department of Oral Biology, State University of New York at Buffalo. d Associate Professor and Director, Division of Oral Facial Prosthetics/Dental Oncology, Department of Otolaryngology, Nebraska Medical Center. a
b
Kim-Pusateri et al
354
Volume 98 Issue 5 There are several problems with making clinical shade selection. Shade selection for dental restorations is affected by many factors, resulting in frequent errors in appearance-matching of restorations to natural teeth. When shades are selected by a human observer, the color-matching process is subjective.1,2 Natural teeth have several characteristics that make shade selection difficult. For example, natural teeth have a double-layering effect caused by enamel translucency and dentin opacity.3 Factors affecting human perception of color, such as lighting conditions and human physiologic variables, cause limitations in traditional visual shade selection.4 The increased emphasis on dental esthetics in recent years has driven the search for a more accurate shade-replication procedure than the present methods provide. Currently available electronic shade-matching devices are spectrophotometers, colorimeters, digital color analyzers, or combinations of these. Spectrophotometers are useful in the measurement of surface color. A prism disperses white light from a tungsten-filament bulb in the spectrophotometer into a spectrum of wavelength bands between 10 and 20 nm.5 The amount of light reflected from a specimen is measured for each wavelength in the visible spectrum. Of all devices, a spectrophotometer is the most accurate for absolute color measurement. These instruments have a longer working life than colorimeters and are unaffected by object metamerism.5,6 Colorimeters are useful in quantifying color differences between specimens. These devices measure tristimulus values according to CIE illuminant and observer conditions.7 Colorimeters use photodiode filters to control light reaching the specimen. The light reflected from the specimen is then measured by a detector. Colorimeters are easier to use and are less expensive than spectrophotometers. However, repeatability may be poor due to aging of filters, and object metamerism can be a chal-
1 ShadeScan (Cynovad) device. lenge to their accuracy.6 ShadeScan (Cynovad, Saint-Laurent, Canada) (Fig. 1) combines a colorimeter with digital imaging. It consists of a handpiece with a liquid crystal display screen which can be seated in a base unit, allowing autocalibration. It contains a halogen light source and uses 45/0 measuring geometry. The device can be used under any lighting condition because the halogen light is bright enough to be unaffected by ambient light.6 The downloaded data identifies the shade closest to the measured specimen. The database of the ShadeScan (Cynovad) software system supports the following shade guides: VITA Classical (VITA Zahnfabrik, Bad Sackingen, Germany), VITA 3D-Master (VITA Zahnfabrik), Ivoclar Chromascop (Ivoclar Vivadent, Schaan, Liechtenstein), Shofu Vintage Halo (Shofu Dental, San Marcos, Calif ), Noritake (Noritake Dental Supply Co Ltd, Nagoya, Japan), and Dentsply EsthetX (Dentsply Caulk, Milford, Del). Currently, there are several electronic shade-matching instruments available for clinical use,8,9,10 but there are limited acceptable in vitro models for evaluating their reliability and accuracy. As more shade-measuring de-
The Journal of Prosthetic Dentistry
vices become commercially available, a standard in vitro model would be useful for predicting their success in clinical practice. The purpose of this study was to evaluate the reliability and accuracy of a newly introduced clinical shade-matching instrument using an in vitro model developed for this investigation. The null hypothesis tested was that there is no difference in reliability and accuracy of the shade-matching instrument when measuring 3 different commercial shade guides.
MATERIAL AND METHODS This in vitro evaluation of a shadematching instrument was based on measurements of shade tabs in a simulated oral environment. Using the ShadeScan (Cynovad) instrument, color measurements were made of 3 commercial shade guides: (1) VITA Classical (VITA Zahnfabrik), (2) VITA 3-D Master (VITA Zahnfabrik), and (3) Chromascop (Ivoclar Vivadent). Ten shade guides from each manufacturer were obtained though convenience sampling from 2 institutions. All shade tabs were first cleaned using an ultrasonic cleaning solution (General Purpose Cleaner, nonammoni-
Kim-Pusateri et al
355
November 2007 ated; L&R Mfg Co, Kearny, NJ) for 15 minutes. They were later cleaned using an ultrasonic cleaner for another 15 minutes. Prior to color measurements, shade tabs were cleaned with soap (Ultra Ivory; Proctor & Gamble, Cincinnati, Ohio) and, subsequently, with distilled water. Simulated clinical conditions were created for color measurement of each shade tab. The shade tab to be measured was placed in the middle of a medium gingiva-colored matrix (Shofu Gummy; Shofu Dental Corp) with identically colored shade tabs placed on either side to simulate adjacent teeth (Fig. 2). The shade measurements were made within a black matte box (19.0 cm x 12.7 cm x 9.5 cm) to simulate the oral cavity. A positioning jig was used to accurately reposition the measuring unit for each shade measurement. The shade-matching device was operated according to the manufacturer’s instructions. For the reliability aspect of the study, each shade tab from 1 VITA Classical (VITA Zahnfabrik) shade guide (16 shade tabs), 1 VITA 3DMaster (VITA Zahnfabrik) shade guide (26 shade tabs), and 1 Chromascop (Ivoclar Vivadent) shade guide (20 shade tabs) were measured 10 nonconsecutive times. Identical repeated shade measurements were considered to be reliable whether or not the instrument measurement matched the actual shade tab. The mean reliability for each shade guide and the overall mean reliability of the shade-matching device were calculated. For the accuracy aspect of the study, each shade tab from 30 shade guides (10 VITA Classical; VITA Zahnfabrik, 10 VITA 3D-Master; VITA Zahnfabrik, and 10 Chromascop; Ivoclar Vivadent) were measured once. Accurate measurements were defined as identical shade matches to the shade tab. The accuracy of measurements was based on whether the shade-matching device selected a shade identical to the shade tab measured. The mean accuracy for each
Kim-Pusateri et al
2 Experimental apparatus. shade guide and the overall mean accuracy of the shade-matching device were calculated. Images were downloaded and analyzed with ShadeScan software (ShadeScan Software Version 3.1.0; Cynovad). For reliability, each shade tab was assessed 10 times for each of the 3 shade guides. The percent agreement of each tab was calculated and the 3 guides were compared by 1-way analysis of variance (ANOVA) followed by the Bonferroni multiple comparison procedure.11 Accuracy was assessed by using 10 guides for each of the 3 types of shade guides. The percent correct was calculated for each and analyzed by a 1-way ANOVA followed by the Bonferroni procedure. Ninetyfive percent confidence intervals for mean reliability and mean accuracy were calculated. Overall, a 5% level of significance was used for ANOVA. Multiple comparisons were not significant unless the corresponding P value was less than .0167. All statistical analyses and calculations were completed with statistical software (StatView statistical program version 5.0.1; SAS Institute, Inc, Cary, NC).
RESULTS
A total of 620 color measurements were made for the reliability evaluation (62 shade tabs measured 10 times) and 620 color measurements were made for the accuracy evaluation (620 shade tabs measured
once). The mean reliability for each shade guide is shown in Table I. Measurements of VITA Classical (VITA Zahnfabrik) produced the highest average percent reliability (95.0%), with 3D-Master (VITA Zahnfabrik) slightly (but not significantly) less at 91.2%. Measurements of Chromascop (Ivoclar Vivadent) were significantly less reliable at 76.5%. The analysis of reliability (Table II) revealed significant differences among the 3 shade guides (ANOVA, P=.002). The Fisher’s PLSD (Table III) indicates that 2 guides were significantly different from one other. The result of this multiple comparison procedure is that the mean reliability of measurements of Chromascop (Ivoclar Vivadent) was significantly lower than the other shade guides evaluated. The mean accuracy for each shade guide is shown in Table IV. Measurements of 3D-Master (VITA Zahnfabrik) produced the lowest average percent of accuracy (54.2%), with VITA Classical (VITA Zahnfabrik) slightly higher at 65.0%. Measurements of Chromascop (Ivoclar Vivadent) produced the highest accuracy (84.5%). The analysis of accuracy (Table V) revealed significant differences among the 3 shade guides (ANOVA, P<.001). The Fisher’s PLSD (Table VI) indicates that all 3 guides were significantly different from each other. This multiple comparison procedure demonstrated that the mean accuracy measurements of all 3 shade guides was significantly different.
356
Volume 98 Issue 5
Table I. Mean (SD) for reliability Shade Guide
Count
Mean (SD)
95% Confidence Interval
VITA Classical
16
95.0 (10.95)
(87.65, 100.00*)
3D-Master
26
91.2 (16.33)
(82.56, 99.75)
Chromascop
20
76.5 (19.81)
(64.61, 88.39)
* Actual estimated upper bound is truncated at 100%
Table II. ANOVA for reliability df
Sum of Squares
Mean Square
F
P
Shade guide
2
3666.7
1833.4
6.794
.002
Residual
59
15920.4
269.8
Table III. Fisher’s PLSD for reliability (significance level, 5%) Mean Difference
Critical Difference
P
VITA Classical vs. 3D-Master
3.846
12.862
.464
VITA Classical vs. Chromascop
18.500
13.577
.001
3D-Master vs. Chromascop
14.654
12.039
.004
Shade Guides
Table IV. Mean (SD) for accuracy (n=10) Shade Guide
Mean (SD)
95% Confidence Interval
VITA Classical
65.0 (7.91)
(58.29, 71.71)
3D-Master
54.2 (4.23)
(50.64, 57.83)
Chromascop
84.5 (13.63)
(72.93, 96.07)
Table V. ANOVA for accuracy df
Sum of Squares
Mean Square
F
P
Shade guide
2
4708.2
2354.1
26.525
<.001
Residual
27
2396.2
88.8
The Journal of Prosthetic Dentistry
Kim-Pusateri et al
357
November 2007
Table VI. Fisher’s PLSD for accuracy (significance level, 5%) Mean Difference
Critical Difference
P
VITA Classical vs. 3D-Master
10.769
10.754
.017
VITA Classical vs. Chromascop
–19.500
10.754
<.001
3D-Master vs. Chromascop
–30.269
10.754
<.001
Shade Guides
DISCUSSION The null hypothesis that there are no differences in reliability among the 3 types of shade guides measured was rejected (ANOVA, P=.002) (Table II). The mean reliability of measurements of Chromascop (Ivoclar Vivadent) shade guides was significantly lower than the other shade guides. The null hypothesis that there are no differences in accuracy among the 3 types of shade guides measured was also rejected (ANOVA, P<.001) (Table V). The mean accuracy measurements of all 3 shade guides were significantly different (P<.001). An important concern of electronic shade-matching devices is their ability to reliably and accurately measure tooth color. Reliability refers to the consistency of the output of the device when measuring a given standard. Accuracy refers to the ability of the device to provide a correct color match for a sample. A standard protocol to evaluate the reliability and accuracy of emerging shade-measuring devices would be beneficial. Since there is no established standard to compare measurements, it has been proposed that shade guides serve as the standard for color measurement.12 In the present study, an in vitro model was developed using shade tabs in a simulated oral environment for the evaluation of electronic shade-matching devices. The use of shade tabs for the target sample offers several advantages. Shade tabs are representative of natural teeth in terms of size, contour, and color. Shade tabs are also readily available and reproducible, assuming an insignificant variation in shade guides from the same manufacturer.
Kim-Pusateri et al
Other substrates with more clinical relevance could have been selected for measurement, such as extracted teeth or ceramic restorations. For the purpose of this study, it was assumed that all shade tabs that were selected represented the variation of the shade guides available from the manufacturers. Shade tabs may vary among shade guides from the same manufacturer, and this was a consideration in measuring the 10 sets of selected shade guides. Inherent differences among the shade tabs may be reflected as incorrect color measurements. Additionally, the irregular surface of a shade tab may affect the spectral reflectance of the shade-matching device, and consequently, the output.13 These potential variations were not evaluated in this study. In this evaluation, reliability and accuracy were the 2 objective measures evaluated. Reliability was defined as the percentage of identical repeated shade measurements, whether or not the measurement matched the actual shade tab. There was no significant difference between VITA Classical (VITA Zahnfabrik) and VITA 3D-Master (VITA Zahnfabrik) shade guides in terms of reliability (P=.4641). However, the reliability between VITA Classical and Chromascop (Ivoclar Vivadent) (P=.001), and VITA 3D-Master and Chromascop (P=.004), was statistically different. Chromascop shade tabs are smaller than both VITA 3D-Master and VITA Classical shade tabs. When Chromascop shade tabs were placed in the Shofu Gummy, spaces existed on either side of the target shade tab which were not present with VITA 3D-Master or VITA Classical shade tabs. This may have
influenced the reliability of ShadeScan (Cynovad) in the measurement of Chromascop shade tabs and is a limitation of the study. For the accuracy evaluation of ShadeScan (Cynovad), accurate measurements were identified as identical shade matches to the shade tab. There were significant differences among the 3 shade guides in terms of accuracy (Table IV). The wide range in accuracy (54.2% to 84.5%) illustrates that variation among shade tabs from different manufacturers influences the accuracy of shade measurements. Variation among natural teeth would be expected in a clinical setting, lending support for the use of multiple shade guides when performing an in vitro evaluation of electronic shadematching devices. The ease of use and degree of technique sensitivity were not evaluated in this study. These factors are not as important under the controlled conditions of this study, but may be important under clinical conditions. For example, holding this device steady is more difficult intraorally compared to use of a positioning jig. Movements made during shade measurement may impair accurate shade measurement. When tooth shade is measured by an electronic shade-matching device, color is expressed in terms of standards established by manufacturers of dental shade guides. The shade tab is therefore the communication tool between dentists and dental laboratory technicians regardless of whether tooth shade is measured visually or with an electronic device. Successful fabrication of a restoration requires the technician to match the color of the definitive ceramic restoration to
358
Volume 98 Issue 5 the prescribed shade tab. However, an imperceptible variation in color matching between a natural tooth and shade tab may become significant if the restoration varies further from the shade tab. Shade tabs serve as an intermediate step and therefore introduce 2 possibly conflicting sources of error – by the dentist in shade selection and by the technician in shade replication. This study does not allow for conclusions to be made regarding the clinical performance of the device tested, since no measurements were made on natural teeth. There are several factors that may influence accuracy and reliability of this device clinically. Unlike the shade tabs used in this study, natural teeth have variable surface texture and anatomic variations that may influence shade measurement. As discussed previously, natural teeth have a 3-dimensional double-layering effect and varying degrees of translucency compared to dental guides.3 Additional in vivo studies are needed before a prediction of clinical performance can be made. Only 1 instrument was evaluated in this study. To determine the overall usefulness of this data, other instruments are being evaluated in an ongoing study following this protocol.
CONCLUSIONS Within the limitations of this study, the following conclusions were made:
1. There was no significant difference in reliability measurements of VITA Classical (95.0%) and VITA 3D-Master (91.2%) shade guides (P=.4641). 2. Significant differences were found with reliability measurements of VITA Classical (95.0%) and Chromascop (76.5%) shade guides (P=.001), as well as VITA 3D-Master (91.2%) and Chromascop (76.5%) shade guides (P=.004). 3. The mean reliability of measurements of Chromascop (Ivoclar Vivadent) shade guides was significantly lower than the other shade guides. 4. Accuracy measurements of VITA 3D-Master produced the lowest average percent accuracy (54.2%), with VITA Classical slightly higher at 65.0%. Measurements of Chromascop produced the highest accuracy (84.5%). 5. The mean accuracy measurements of all 3 shade guides were significantly different (P<.001).
REFERENCES 1. Culpepper WD. A comparative study of shade-matching procedures. J Prosthet Dent 1970;24:166-73. 2. Geary JL, Kinirons MJ. Colour perception of laboratory-fired samples of body-coloured ceramic. J Dent 1999;27:145-8. 3. O’Brien WJ. Double layer effect and other optical phenomena related to esthetics. Dent Clin North Am 1985;29:667-72. 4. Carsten DL. Successful shade matching – what does it take? Compend Contin Educ Dent 2003;24:175-8, 180, 182. 5. Berns RS. Billmeyer and Saltzman’s principles of color technology. 3rd ed. New York: John Wiley & Sons Inc; 2000. p.
88-92. 6. Paravina RD, Powers JM. Esthetic color training in dentistry. St. Louis: Mosby; 2004. p. 17-28, 169-170. 7. CIE (Commission Internationale de l’Eclairage). Colorimetry, official recommendations of the International Commission on Illumination, Publication CIE No.15:2004: Colorimetry, 3rd ed. 8. Paul S, Peter A, Pietrobon N, Hammerle CH. Visual and spectrophotometric shade analysis of human teeth. J Dent Res 2002;81:578-82. 9. Yap AUJ, Sim CPC, Loh WL, Teo JH. Human-eye versus computerized color matching. Oper Dent 1999;24:358-63. 10.Okubo SR, Kanawati A, Richards MW, Childress S. Evaluation of visual and instrument shade matching. J Prosthet Dent 1998;80:642-8. 11.Fleiss JL. The design and analysis of clinical experiments. New York: John Wiley & Sons Inc; 1986. p. 8-13, 46-88. 12.Dozic A, Kleverlaan CJ, El-Zohairy A, Feilzer AJ, Khashayar G. Performance of five commercially available tooth color-measuring devices. J Prosthodont 2007;16:93-100. 13.Seghi RR. Effects of instrument-measuring geometry on colorimetric assessments of dental porcelains. J Dent Res 1990;69:1180-3. Corresponding author: Dr Seungyee Kim-Pusateri State University of New York at Buffalo Department of Restorative Dentistry School of Dental Medicine 3435 Main St. Squire Hall Room 222 Buffalo, NY 14214 Fax: 716-829-2440 E-mail: [email protected] Acknowledgements The authors thank Cynovad (Saint-Laurent, Canada), Ms April Logue, former Research Associate at Section of Restorative and Prosthetic Dentistry, for assistance, and Corey E. Valentine for assistance with ShadeScan measurement. Copyright © 2007 by the Editorial Council for The Journal of Prosthetic Dentistry.
Availability of Journal Back Issues As a service to our subscribers, copies of back issues of The Journal of Prosthetic Dentistry for the preceding 5 years are maintained and are available for purchase from Elsevier Inc until inventory is depleted. Please write to Elsevier Inc, Subscription Customer Service, 6277 Sea Harbor Dr, Orlando, FL 32887, or call 800-654-2452 or 407-345-4000 for information on availability of particular issues and prices.
The Journal of Prosthetic Dentistry
Kim-Pusateri et al