Академический Документы
Профессиональный Документы
Культура Документы
available at www.sciencedirect.com
a r t i c l e i n f o a b s t r a c t
Article history: Objectives. The objectives were to measure the difference in the color and color parame-
Received 13 February 2006 ters of natural teeth measured by a tristimulus colorimeter (CM, used as a reference) and
Accepted 15 November 2006 Shade Vision System (SV), and to determine the influence of color parameters on the color
difference between the values measured by two instruments.
Method. Color of 12 maxillary and mandibular anterior teeth was measured by CM and SV
Keywords: for 47 volunteers (number of teeth = 564). Color parameters such as CIE L* , a* and b* values,
Tooth color chroma and hue angle measured by two instruments were compared. Chroma was calcu-
1/2
Colorimeter lated as C∗ab = (a∗2 + b∗2 ) , and hue angle was calculated as h◦ = arctan(b* /a* ). The influence
Shade Vision System of color parameters measured by CM on the color difference (E∗ab ) between the values
Color difference measured by two instruments was analyzed with multiple regression analysis (˛ = 0.01).
Results. Mean E∗ab value between the values measured by two instruments was 21.7 (±3.7),
and the mean difference in lightness (CIE L* ) and chroma was 16.2 (±3.9) and 13.2 (±3.0),
respectively. Difference in hue angle was high as 132.7 (±53.3)◦ . Except for the hue angle, all
the color parameters showed significant correlations and the coefficient of determination
(r2 ) was in the range of 0.089–0.478. Based on multiple regression analysis, the standard-
ized partial correlation coefficient (ˇ) of the included predictors for the color difference was
−0.710 for CIE L* and −0.300 for C∗ab (p < 0.01).
Conclusion. All the color parameters showed significant but weak correlations except for hue
angle. When lightness and chroma of teeth were high, color difference between the values
measured by two instruments was small. Clinical accuracy of two instruments should be
investigated further.
© 2006 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
1. Introduction with a large set of color tabs. Value is determined first, and
chroma is determined next with tabs that are close to the
Color can be described according to the Munsell color space measured value but are of increasing saturation of color. Hue
in terms of hue, value and chroma. When the color is mea- is determined last by matching with color tabs of the value
sured by a visual technique, the color considered is compared and chroma already determined [1]. In 1976, the Commission
∗
Corresponding author. Tel.: +82 2 740 8693; fax: +82 2 740 8694.
E-mail address: ykleedm@snu.ac.kr (Y.-K. Lee).
0109-5641/$ – see front matter © 2006 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.dental.2006.11.008
1308 d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1307–1312
Internationale de l’Eclairage (CIE) refined a color space. CIE L* ferences in measured tooth color depending on measuring
value is a measure of the lightness of an object, CIE a* value instruments; however, there were few studies on the differ-
is a measure of redness or greenness, and CIE b* value is a ence in color and color parameters of teeth measured by a
measure of yellowness or blueness [2]. tristimulus colorimeter and Shade Vision System in the CIE
The color and appearance of teeth is a complex phe- L* , a* and b* scale. The working hypothesis in this study was
nomenon, with many factors such as lighting conditions, that all the color parameters such as CIE L* , a* and b* values,
translucency, opacity, light scattering, gloss and the human chroma and hue angle measured by two instruments showed
perception [3]. The measurement of teeth color remains a significant correlations regardless of the shade of teeth. The
challenge [4]. Therefore, a thorough understanding of appear- objectives of this study were to measure the difference in
ance attributes of natural teeth is required along with new the color and color parameters of natural teeth measured
shade guides and shade-taking instruments to maximize by a tristimulus colorimeter and Shade Vision System, and
shade-matching results [5]. to determine the influence of color parameters on the color
Tooth color is measured by varied methods including difference between the values measured by two instruments.
visual assessment with a shade guide, spectrophotometer,
colorimeter and computer analysis of digital images [3]. Due
to inter-human differences in the perception of color, visual 2. Materials and methods
shade assessment of teeth is lacking standardization that may
be improved by the use of a spectrophotometer. Spectropho- 2.1. Tooth color measurement
tometric shade analysis of natural teeth was more accurate
and more reproducible compared with human shade assess- Total of 47 volunteers, older than 19-year-old, were included
ment (83.3% compared to 26.6%) [6]. Assessment of tooth color in this study. Approval was obtained from the institutional
change was performed with a small-area colorimeter, and the review board and informed consent was obtained from each
standard deviation for CIE L* value was 0.22 [7]. A study on patient. Through clinical examinations, their general and
the accuracy of a colorimeter was concluded that this instru- oral health conditions were checked. They did not have
ment could be used as a tool for the quantitative assessment caries, abraded lesions or restorations in the 12 maxillary and
of the gradual change in the shade of teeth [8]. However, mandibular anterior teeth. Their periodontal conditions were
it was reported that the accuracy of a new colorimeter in fair to good. Eight males and 39 females were included, and
matching porcelain shade guide teeth was only slightly better their mean age was 29.5 (±5.2) and 29.0 (±6.8), respectively.
than visual matching (50% in colorimeter and 48% in visual Color of 12 maxillary and mandibular anterior teeth was
matching) [9]. The disadvantages of using a colorimeter for measured by a tristimulus colorimeter (CM: Chroma Meter
measuring tooth color include: the instrument is designed to CR 321, Minolta, Osaka, Japan) and Shade Vision System (SV:
measure flat surfaces, and small aperture colorimeter is prone X-rite, Grandville, MO, USA). When CM was used, individual
to significant edge-loss effects [10–12]. positioner was not fabricated because the total number of
In addition to tristimulus colorimeters, a new clinical col- teeth was too many (564 teeth), and also it was intended to
orimeter system, Shade Vision System (X-rite, Grandville, MO, obtain the tooth color affected by the surrounding tissue. To
USA), was introduced. Shade Vision System is a commer- exclude the influence of direct sunlight, all the measurements
cially available shade-taking system that provides an accurate were performed at a unit chair receiving no direct sunlight
colored ‘contour map’ image of the tooth. It is essentially a between 5 and 6 p.m., and one dentist measured the color
colorimeter that utilizes image-grabbing technology, and com- based on the same measurement protocol. The aperture diam-
prises a handheld measuring instrument that is used to scan eter of the colorimeter was 3 mm, and the illuminating and
the tooth surface together with a docking station linked to a viewing configuration was 45◦ illumination/0◦ viewing angle.
computer and associated software [13]. Correlations on the After calibration using a white standard tile, aperture head of
measured color of natural teeth before and after bleaching the colorimeter was contacted to the center of a tooth. Mea-
using a standard clinical shade guide, Shade Vision System surements were repeated three times. When SV was used,
and a colorimeter were determined [13–15]; however, they all the measurements were scheduled between 5 and 6 p.m.
used shade guide unit (SGU) instead of CIE L* , a* and b* values to exclude the influence of direct sunlight, and one dentist
in Shade Vision System. They regarded E∗ab values measured measured the color based on the same measurement pro-
by a colorimeter of 1.76–4.17 as 0 SGU change (no change) with tocol. Aperture head was contacted to the center of a tooth.
Shade Vision System. Measurements were repeated three times.
The result of investigations on the relationship between the
color perceived by human observers and the color measured 2.2. Comparison of tooth color measured by two
by a colorimeter was inconclusive [16]. There was significant instruments
correlation between instrumental colors and human obser-
vations [10,17], whereas there was no significant correlation In both instruments, the CIELAB system allows color speci-
between them [9,18]. Colorimeter measurements were com- fication within a three-dimensional color space. Differences
pared with spectrophotometer readings [17,18]. Though the in CIE L* , a* and b* values measured by two instruments
accuracy of a colorimeter on the tooth color measurement were calculated, and the color difference between the color
is inconclusive, this instrument has been used for a long coordinates measured by two instruments was calculated
2 2 2 1/2
time; therefore, can be regarded as a reference to a new color as: E∗ab = [(L∗s − L∗c ) + (a∗s − a∗c ) + (b∗s − b∗c ) ] , in which sub-
measuring instrument. There have been studies on the dif- script s and c indicates SV and CM, respectively. Since the color
d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1307–1312 1309
CM SV CM SV
3. Results
Table 2 – Correlation between each pairs of CIE L* , C∗ab and h◦ values measured by CM or SV based on all the teeth
measured
Y X Measured by CM Measured by SV
All the coefficients of determination (r2 ) were significant at the level of 0.01.
(ˇ) for CIE L* and C∗ab was −0.710 and −0.300, respectively
(p < 0.01).
Correlation between each pairs of CIE L* , C∗ab and h◦ val-
ues measured by CM or SV based on all the teeth measured
are presented in Table 2. When measured by CM, there were
significant correlations between CIE L* and h◦ , and C∗ab and
h◦ . When measured by SV, there were significant correlations
between CIE L* and C∗ab , and CIE L* and h◦ . The coefficient of
determination increased when measured by SV.
4. Discussion
objective measurements of tooth color, it can be tedious to use As a result, shade determination by visual means was incon-
with a custom alignment instrument [4]. The data measured sistent. Accuracy of a new colorimeter in matching porcelain
by this instrument are inconsistent with the commonly used shade guide teeth was only slightly better (50% in colorime-
visual instruments such as Vitapan Classical shade guide, ter and 48% in visual matching). Though the manufacturer of
although in many cases the general trends are similar. It is the colorimeter was different from that used in the present
also questionable whether the small area measured repre- study, the aperture size was similar (3 mm diameter in the
sents the color of the whole tooth adequately. Consequently, present study). Therefore, the accuracy of the colorimeter used
at present the data obtained with this instrument alone do not in the present study was concerned. The shape of shade guide
appear to be adequate for determining tooth color change in is similar to that of natural tooth; therefore, no flat surface for
whitening research, although the quantitative measurements color measurement. This should have influenced the measur-
may be useful as supplemental or supportive data [4]. In the ing accuracy.
present study, individual positioner could not be fabricated Color distribution in three regions in a sample of 95
because the total number of teeth was too many (564 teeth). extracted human teeth was determined using a recording
Since there have been no report on the accuracy of SV, CM was spectrophotometer. The mean L* , a* and b* values were 72.6, 1.5
regarded as the reference. Compared with the dada measured and 18.4 for gingival, 72.4, 1.2 and 16.2 for middle, and 71.4, 0.9
by SV, there was significant difference, especially in hue angle. and 12.8 for incisal [20]. In the present study, mean L* , a* and
But at present, there are no guidelines for the accuracy of CM b* value was 57.8, −1.0 and 6.7, respectively, with CM. When
and SV, in other words which instrument measures the color measured by SV, the mean L* , a* and b* value was 74.0, 5.0 and
accurately. Further study is recommended. 19.4, respectively. In the present study, the color of only cen-
There appear to be marked differences in the CIE L* , a* tral point (3 mm diameter) of tooth was measured by CM and
and b* values of tooth obtained by CM compared with other the color coordinates were smaller than those measured in the
systems, such as other colorimeters, spectrophotometers and previous study of O’Brien et al. [20]. But the values measured
digital imaging system that are also capable of generating the by SV were similar to those of the previous study of O’Brien et
same parameters. In general, the CIE L* , a* and b* values of al. [20].
tooth obtained with this instrument tend to be lower than There have been many studies on the percepti-
those obtained with other systems (CIE L* , a* and b* values ble/acceptable limit of color difference. A color difference
were 40–60, −1.5–0.5 and 2–12 with CM, and 65–75, −1.5–10 value of greater than 2 E∗ab unit was perceived by the 100%
and 12–23 with other instruments) [4]. Theoretically, the mea- observers, and E∗ab units between 1 and 2 were not perceived
surements should be the same for the same color as they are infrequently [21]. Thresholds for perceptibility (mean; 0.4 E∗ab
all based on the three-dimensional coordinates of the CIELAB units) were significantly lower than thresholds for accept-
color space. One of the possible factors that may contribute to ability (mean; 1.7 E∗ab units) [22]. As to the acceptability, 50%
this difference is the small (3 mm in diameter) measurement acceptability was approximately 1 E∗ab unit [23]. E∗ab values
aperture of this instrument. A study using a spectrophotome- of 3.3 were unacceptable in 50% of the observers [24].
ter and a spectroradiometer found that the CIE L* , a* and b* Discoloration model was assessed by a shade guide, Shade
values obtained from extracted human teeth decreased with Vision System and a colorimeter. In the group of no change
smaller-size measuring windows [12]. The trend for CIE a* and (shade guide unit change = 0) with a shade guide and Shade
b* value measurements was similar, indicating that tooth color Vision System, color changes were observed when measured
measurements using a small window tend to shift toward by a colorimeter (E∗ab = 3.90–4.17, L* = 3.22–3.58, a* = − 0.25
green and blue color coordinates in the CIELAB color space, to −0.07, b* = − 0.40 to 0.56) [13]. In this case, E∗ab value was
which was explained by the wavelength-dependent edge loss in the unacceptable range even when the highest threshold
[12]. In the present study, color coordinates measured by CM value claimed by Ruyter et al. [24] (E∗ab = 3.3) was adopted;
was smaller than those with SV, and the differences were however, the change in color measured by SV was reported
16.2 for L* , 6.0 for a* and 12.7 for b* . This result could be as no change. Bleaching effect was assessed with the same
explained by the wavelength-dependent edge loss. In addi- three instruments used in the previous study of Sulieman et
tion, CIE a* values measured by CM distributed from −5.1 to al. [13]. In the group of no change with a shade guide and Shade
4.0, whereas those with SV distributed from 1.6 to 9.8. The Vision System, color changes were observed when measured
difference in the range of distribution was not so big (9.1 and by a colorimeter (E∗ab = 2.6 − 3.2) [14]. If the E∗ab value of 3.3
8.2); however, negative value with CM influenced hue angle is adopted as the threshold acceptable value [24], these values
significantly. Additionally, correlation between each pairs of were in the range of acceptable value. But except the report
CIE L* , C∗ab and h◦ values measured by CM or SV showed clear of Ruyter et al. [24], others proposed smaller E∗ab value for
difference, which implied that tooth color was influenced by acceptability (E∗ab = 1 − 1.7). If these criteria are adopted, the
the measuring instrument. Although it has not been con- color difference which a shade guide and Shade Vision System
firmed which instrument reflects the real color of teeth, the resulted in no change was unacceptable. The effect of hydro-
discrepancy in tooth color by the instrument should be studied gen peroxide on the shade change of teeth was determined
further. with the same three instruments used in the previous study
Ability of a new computerized colorimeter equipped with [13]. In the group of no change with a shade guide and Shade
a positioning guide and a simple visual test to match ceramic Vision System, small changes were observed by a colorime-
shade guide teeth was evaluated. The colorimeter comes with ter (E∗ab = 1.76, L* = 0.49, a* = 0.21, and b* = 1.03) [15]. If the
a 3 mm × 4 mm aperture that attaches to the bottom of the E∗ab value of 1–1.7 was regarded as the threshold acceptable
unit, which can be placed directly on an anterior tooth [9]. value [22,23], the color difference which a shade guide and
1312 d e n t a l m a t e r i a l s 2 3 ( 2 0 0 7 ) 1307–1312
Shade Vision System resulted in no change was unacceptable. [9] Okubo SR, Kanawati A, Richards MW, Childress S. Evaluation
Therefore, the accuracy of SV was concerned also. Based on of visual and instrument shade matching. J Prosthet Dent
the previous studies [13–15], CM was more accurate than SV. 1998;80:642–8.
[10] van der Burgt TP, Ten Bosch JJ, Borsboom PCF, Kortsmit
But it is not clear whether CM really reflect the color of natural
WJPM. A comparison of new and conventional methods for
tooth. Further study is recommended. quantification of tooth color. J Prosthet Dent 1990;63:
155–62.
[11] ten Bosch JJ, Coops JC. Tooth color and reflectance as related
5. Conclusions
to light scattering and enamel hardness. J Dent Res
1995;74:374–80.
Within the limitations of the present study, except for the [12] Bolt RA, ten Bosch JJ, Coops JC. Influence of window size in
hue angle, color parameters measured by CM and SV showed small-window color measurements, particularly of teeth.
significant correlations and the values measured by SV were Phys Med Biol 1994;39:1133–42.
higher than those with CM. But the coefficient of determina- [13] Sulieman M, Addy M, Rees JS. Development and evaluation
tion was not high. Some color parameters of teeth influenced of a method in vitro to study the effectiveness of tooth
bleaching. J Dent 2003;31:415–22.
the color difference between the values measured by two
[14] Sulieman M, MacDonald E, Rees JS, Addy M. Comparison of
instruments. Clinical accuracy of two instruments should be three in-office bleaching systems based on 35% hydrogen
investigated further. peroxide with different light activators. Am J Dent
2005;18:194–7.
references [15] Sulieman M, Addy M, Macdonald E, Rees JS. The bleaching
depth of a 35% hydrogen peroxide based in-office product: a
study in vitro. J Dent 2005;33:33–40.
[16] Tung FF, Goldstein GR, Jang S, Hittelman E. The repeatability
[1] Craig RG, Powers JM. Restorative dental materials. 11th ed. of an intraoral dental colorimeter. J Prosthet Dent
Mosby; 2002. p. 41. 2002;88:585–90.
[2] CIE (Commission Internationale de l’Eclairage). [17] Johnston WM, Kao EC. Assessment of appearance match by
Colorimetry—Technical Report. CIE Pub. No.15, 2nd ed. visual observation and clinical colorimetry. J Dent Res
Vienna, Austria: Bureau Central de la CIE, 1986 (corrected 1989;68:819–22.
reprint 1996), p. 35–6. [18] Goldstein GR, Schmitt GW. Repeatability of a specially
[3] Joiner A. Tooth color: a review of the literature. J Dent designed intraoral colorimeter. J Prosthet Dent
2004;32(Suppl. 1):3–12. 1993;69:616–9.
[4] Li Y. Tooth color measurement using Chroma Meter: [19] Norman GR, Streiner DL. Biostatistics. St. Louis: Mosby; 1984.
techniques, advantages, and disadvantages. J Esthet Restor p. 100–16.
Dent 2003;15(Suppl. 1):S33–41. [20] O’Brien WJ, Hemmendinger H, Boenke KM, Linger JB, Groh
[5] Brewer JD, Wee A, Seghi R. Advances in color matching. Dent CL. Color distribution of three regions of extracted human
Clin North Am 2004;48:341–58. teeth. Dent Mater 1997;13:179–85.
[6] Paul S, Peter A, Pietrobon N, Hammerle CH. Visual and [21] Seghi RR, Hewlett ER, Kim J. Visual and instrumental
spectrophotometric shade analysis of human teeth. J Dent colorimetric assessments of small color differences on
Res 2002;81:578–82. translucent dental porcelain. J Dent Res 1989;68:1760–4.
[7] Rustogi KN, Curtis J. Development of a quantitative [22] Douglas RD, Brewer JD. Acceptability of shade differences in
measurement to assess the whitening effects of two metal ceramic crowns. J Prosthet Dent 1998;79:254–60.
different oxygenating agents on teeth in vivo. Compend [23] Kuehni RG, Marcus RT. An experiment in visual scaling of
Suppl 1994;17:S631–4. small color differences. Color Res Appl 1979;4:83–91.
[8] Amaechi BT, Higham SM. Development of a quantitative [24] Ruyter IE, Niler K, Moller B. Color stability of dental
method to monitor the effect of a tooth whitening agent. J composite resin materials for crown and bridge veneers.
Clin Dent 2002;13:100–3. Dent Mater 1987;3:246–51.