WORKSHOP INTERNACIONAL PROGRAMA DE PÓS‐GRADUAÇÃO EM BIOMATERIAIS E BIOLOGIA ORAL CONVIDA: 

 

 

 

 

 

DATA: 07 DE OUTUBRO DE 2011.  LOCAL: Faculdade de Odontologia da USP.  HORÁRIO: 09:00 às 13:00.  O workshop será gratuito.  Não há necessidade de inscrição.  Contato: Prof. Paulo Francisco Cesar (paulofc@usp.br) 

 

Palestrante: Allyson A Barrett  Instituição: Center for Dental Biomaterials, College of Dentistry, University of Florida  Título do Workshop: “Combining Art and Science for Visual Acuity and Dental Shade Matching”

Resumo do Workshop: This session is designed to test colour vision acuity – both the ability to recognize and discriminate colours ‐ using a standardized test and objective exercises with restorative dental materials. The idea is to give the dentist the knowledge they need for shade evaluation in the clinic on a daily, ongoing basis since the eye remains the reliable, clinical tool of choice. Colour recognition and therefore,  shade matching  ability,  can  be  enhanced  and  improved.    You must  first  have  objective  knowledge  of  your  own individual ability and some understanding of  the subjectivity  in colour perception.    In years of colour science  research within Dental Biomaterials at the University of Florida we have developed particular exercises which have been presented at both AADR and IADR conferences. Understanding effects of visual phenomena, the three‐dimensional colour system that provides the basis for  instrumental colour determination, and  features of  intra‐oral  instrumental shade analysis are part of  this session. You will have a reasonably accurate  idea of your own colour vision (quantitative score) and  learn the difference between colour vision acuity and colour discrimination ability. Visual colour perception ability can be learned and improved! - Objective Visual Acuity Exercises: • Isochromatic Plates   (for the red‐green colour axis vision); • Discrimination Pairs   (Dental ceramic discs fabricated specifically for a wide range of colour differences, Delta E values, allow determination of  one’s colour discrimination ability). 

‐ Organizational effects of Shade guide selections  ‐ Visual Perception Images (designed to “teach the eye” discriminatory nuances) 

Biografia resumida: Interest in colour comes to Allyson A. Barrett through years of studying painting with recognized artists from different countries.  Going from Arts to Dental Sciences, her early work was with LL Hench in Bioglass and, then for many years, with KJ Anusavice at the Center for Dental Biomaterials, College of Dentistry, University of Florida. Her ceramics research  includes optical, thermal and mechanical property characterization.   She  teaches colour  to  the UF dental students,  residents, continuing education  for practicing clinicians, on cruise ships and to private professional organizations. She feels that the eye is still the primary clinical “tool” and that knowing one’s own visual acuity is paramount to aesthetic selections.  Colour is an integral part of our daily lives! 

Sugestão de leitura:  Influence of tab and disk design on shade matching of dental porcelain. Barrett AA, Grimaudo NJ, Anusavice KJ, Yang 

MC. J Prosthet Dent. 2002 Dec;88(6):591‐7.  Visual and instrumental agreement in dental shade selection: three distinct observer populations and shade matching 

protocols. Della Bona A, Barrett AA, Rosa V, Pinzetta C. Dent Mater. 2009 Feb;25(2):276‐81. 

Influence of tab and disk design on shade matching of dental porcelain

Allyson A. Barrett,a Nicholas J. Grimaudo, DMD, MS,b Kenneth J. Anusavice, PhD, DMD,c andMark C. K. Yang, PhDd

College of Dentistry, University of Florida, Gainesville, Fla.

Statement of problem. Given the complexity of tooth color, the variations of shade within each tooth, andtranslucency, it is difficult to view only one small area and select a shade match for restorations.Purpose. This study tested the effect of specimen design on porcelain shade matching, hypothesizing that flatdisks would be matched to one another with more accuracy than tooth-shaped tabs to tabs.Material and methods. All testing was conducted in a Macbeth SpectraLight booth with D65 illumination.Seventy-three senior dental students (25 women and 48 men; mean age, 27 years) were asked to match selectedVita porcelain disks and Vita shade tabs to like specimens. The design order, namely matching tabs or disks first,was alternated for each observer. The specimens were handed to the observer individually. No time limit formatching was imposed, although each observer was given explicit instructions related to the observation andhandling of the specimens. Upon completion of the matching exercises, each student received his or herstandardized test results and reviewed the matching results. The time for testing and review was approximately 20minutes per observer. An analysis of variance, with gender and order as 2 factors that could affect matching scores,was performed (P !.05).Results. The mean matching scores were 78.4% for disks and 73.6% for tabs (P".119). Female observersmatched 76.5% of the disks and 77.5% of the tabs, whereas male observers matched 79.4% of the disks and 71.6%of the tabs (P".054). Matching disks before tabs yielded equivalent levels of shade matching (disks, 77.6%; tabs,77.1%). When tabs were matched first, the scores were as follows: disks, 79.8%, and tabs, 67.3% (P".010).Conclusion. Within the limitations of this study, there was no significant difference in shade-matchingaccuracy between the 2 shapes, although the order of design matching resulted in a difference in shade-matchingability. When tabs were matched first and disks second, improved matching was evident on the second test. Thereverse was not true; no learning was demonstrated when the tabs were matched after the disks. (J Prosthet Dent2002;88:591-7.)

CLINICAL IMPLICATIONS

In this study the gender of the shade matcher had no significant relevance to dental shadeselection ability, in contrast to previous reports in the dental literature. Color-blind individualsmay be able to discriminate adequately for acceptable dental shade matching. All dental per-sonnel may benefit from an objective knowledge of their own color-/shade-matching ability,particularly because proficiency in shade matching can be learned and developed through mul-tiple objective exercises.

Color matching and shade matching are an essen-tial aspect of restorative and esthetic dentistry andhave been accentuated by the recent social emphasis,attached to esthetic dentistry. Although there have

been developments in restorative dental materials andinstrumentation over the years, intraoral shadematching has not changed significantly since the ini-tial dental studies by Clark.1 Intraoral shade matchingis still dependent upon each clinician’s visual discrim-ination and matching abilities. An objective knowl-edge of one’s color vision is essential to optimize theesthetics of patients’ prostheses. Learning to matchshades, characterize restorations, and alter shades isdependent on one’s ability to perceive and discrimi-nate colors and differences. Color screening of dentalprofessional personnel is not typically done in NorthAmerican dental schools, even though it is a factor indaily restorative dentistry. The fact that it is possibleto improve one’s color perception2 emphasizes theneed for color instruction in dental education.

A preliminary report of this research was published in abstract form inJ Dent Res 1999;78:80 and presented at the 77th General Session ofthe International Association of Dental Research, March 1999, Van-couver, British Columbia, Canada.

This study was supported by NIH-NIDCR grant DE06672-16.aResearch support staff, Department of Dental Biomaterials.bDirector of Quality Assurance and Assistant Professor, Department

of Operative Dentistry.cAssociate Dean for Research; Professor and Chair, Department of

Dental Biomaterials.dProfessor, Department of Statistics.

DECEMBER 2002 THE JOURNAL OF PROSTHETIC DENTISTRY 591

The few studies of visual color testing published inthe dental literature usually describe limited testing.Moser et al3 screened 670 participants at the 1981 ses-sion of the American Dental Association. Testing forred-green deficiency only, they found that 9.9% of thedental professionals exhibited some color vision defi-ciency. Serious color vision deficiency was found in 2.8%of that dentally related population. Rawlinson4 foundthat 2 of 6 dental students with defective color visionwere unable to see yellow or blue, a parameter relevantto dental shades that often goes untested. Many of thestudents were unaware of their visual limitations. Con-sequently, the findings supported the continued use ofvisual screening to identify, as well as to encourage,those with color vision deficits to seek professional treat-ment at an early opportunity. O’Keefe et al5 and Wassonand Schuman6 advocated color vision screening as wellas improved color education for dentists and dental stu-dents. Addressing the primary elements of color,O’Keefe et al noted that shade matching is the weak linkin esthetic dentistry. Wasson and Schuman advocatedproduct development that may help in shade selection,as they found that 9.3% of 150 subjects screened forred-green color vision had defective color vision.

Because the tooth is not a homogeneous surface,shade matching is considered to be very difficult. Assess-ing color-matching ability with the use of natural teethor ceramic shade tabs is also considered to be difficult, asthere are distinct areas of varying color and translucencywithin a single tooth. Determining specifically whicharea an observer is viewing for shade-matching purposesis complicated. van der Burgt et al7 used a perforatedshield to consistently expose only a limited section of thetooth for viewing. Two viewing methods were used, anda difference in interobserver agreement was found. Theinterobserver differences were somewhat greater thanthe intraobserver differences. The authors concludedthat even under the same conditions, visual and instru-mental color determinations are subject to large errors.

The eye is very discriminating, detecting inhomoge-neities and variations on small surfaces. Instrumentssuch as colorimeters and spectrophotometers are de-signed to measure color by integrating stimuli on a des-ignated flat surface. These instruments then print 3 nu-merical coordinates from a 3-dimensional color spaceestablished by the Commission on Illumination.8 Theinstrumental apertures are usually 3 mm, 5 mm, or8 mm in diameter. The smaller apertures (3 and 5 mm)generally result in color coordinate shifts due to “edgeloss.”9 A colorimeter with a minimum aperture of 8 mmis considered to be more reliable in repeat measure-ments, whereas human beings are less consistent fromobservation to observation, even under the same view-ing conditions. If the surface being “measured” has anysurface or color irregularities or is composed of multiplecolors, it will all be integrated within the aperture area,

and only parameters for 1 specific color will be gener-ated.

Clarke10 noted that data from such irregularities mayresult in mistaken color identification. Instruments arealso subject to mechanical problems. Goodkind andSchwabacher11 found it necessary to eliminate data from1 of 4 colorimeters used in an intraoral study because ofinstrumental problems. Consequently, instrumentationcan be considered accurate and effective for measuringflat homogenous materials and for making fine colordistinctions that may not necessarily be visually detect-able.12 Teeth do not fit this description. However, theeye, which is an excellent detector, does observe/recordnonhomogeneity. Okubo et al13 reported that a color-imeter was only 50% accurate in matching shade tabs,whereas the matching accuracy of human observers was48%. No significant difference was found between theshade-matching accuracy of the colorimeter and humanobservers. Even with the ability of instruments to makesome fine color distinctions, these findings demonstratethe need for development of perceptual color educationin dentistry.

There is an important distinction between normalcolor vision and visual color discrimination. Color dis-crimination, which is the ability to detect differencesamong colors, may be independent of defective colorvision, a condition in which one is unable to perceive, letalone identify, some colors. Farnsworth14 noted thatsome observers may demonstrate low discriminationability without an imbalance of color perception. Theconverse is also possible: an individual may not be ableto perceive colors (that is, may be color blind), yet thatsame individual may be able to distinguish differences.

Given the complexity of tooth color, the variations ofshade within each tooth, and translucency, it is difficultto view only one small area and select a shade match forrestorations. In an effort to determine the effect of theirregularities of a tooth on dental shade-matching accu-racy, 2 different shapes or designs of dental porcelainwere selected for matching exercises. Dental porcelaindisks with a flat homogeneous surface and porcelaintooth-shaped replicas (shade tabs) simulating naturaltooth variations were used. This study tested the hy-pothesis that observers would match the flat, homoge-neously colored disks to one another more consistentlythan the polychromatic tooth-shaped shade tabs to tabs.

MATERIAL AND METHODSThe observers consisted of 73 senior dental students

(25 women and 48 men) with a mean age of 27 years.Before the dental porcelain shade-matching exercises,each dental observer took 2 standardized tests to assesshis or her chromatic perception relative to both the red-green color axis and the blue-yellow axis. The standardPseudo-Isochromatic plate test (Richmond Products,

THE JOURNAL OF PROSTHETIC DENTISTRY BARRETT ET AL

592 VOLUME 88 NUMBER 6

Boca Raton, Fla.), which screens only for defective red-green vision, was administered first, followed by the Sat-urated 15 Hue Test (Luneau Ophtalmologie, Pads,France). The latter is considered to be a more accuratetest for even moderate color anomalies on both chro-matic axes (red-green and blue-yellow). Ail scores wererecorded for each individual and each test exercise. Noobservers or data were excluded as a result of the stan-dardized scores.

The observers were asked to match the same shapeddental porcelains (disks and tabs) to one another, ac-cording to shade. The exercises were conducted in alight booth with standardized daylight (D65) illumina-tion (Macbeth, Newburgh, N.Y.) against a neutral grayinterior known as Munsell N-7. The booth was the solelight influence in the testing room. No time limit formatching was imposed, although each observer wasgiven explicit instructions related to the observation andhandling of the specimens. Observers were informedthat prolonged viewing could result in foveal fatigue andthat viewing the neutral gray surroundings could elimi-nate the deleterious effect.

Two different shapes of Vita dental porcelain (Vi-dent, Brea, Calif.) were used for observer shade match-ing. Specimens included one set of flat porcelain disks(16 # 2 mm) and one set of tooth-shaped tabs. Thedisks were homogeneous, sintered according to themanufacturer’s instructions, and polished with 1-!malumina abrasive. Considering possible batch variation,the disks for each shade were made from the same con-tainer of porcelain powder. The finished disks were mea-sured with a colorimeter (CR-300; Minolta, Ramsey,N.J.) to ensure acceptable $Es for matching pairs($E ! 1). The second design set comprised commer-cially produced, polychromatic, tooth-shaped shadetabs. The same 8 shades (A1, A2, A3, B2, B3, C1, C2,and D2) were repeated for both the disks and the tabs.The shades were arranged for matching as follows: A1,C1, B2, A2, A3, B3, C2, and D2. This shade arrange-ment did not represent a particular order of value orshade group. Half of the disks were placed in a rowwithin the light booth. Observers were instructed not tomove the disks in this arrangement. The specimen shapeto be matched was handed to the observer individually,in random order. The observer was asked to select theclosest possible match from the specimens within thebooth.

The Vita tab shades were arranged in the same orderas the disks. They were placed within the Vita Lumin tabholder (Vident, Brea, Calif.) by their stainless steel stemswith 1 space between each tab. The same instructionsassociated with disk-to-disk matching were given fortab-to-tab matching. Observers were not allowed to re-arrange the tabs, but they could place the Lumin setholder in their hand. The design configurations (eitherdisks first or tabs first) were alternated with each ob-

server. Upon completion of the matching exercises, eachstudent received his or her standardized test results andreviewed the matching results. Shade/color compari-sons were viewed as was applicable to each observer’sresults so that color distinctions could be noted visuallyand characteristics discussed. The time for testing andreview was approximately 20 minutes per observer.

An analysis of variance, with gender and order as 2factors that could affect matching scores, was performed(P!.05). The interactions between gender and orderwere considered but proved to be insignificant.

RESULTSAlthough the disks were correctly matched more fre-

quently than the tabs, the difference did not reach sta-tistical significance. The overall disk-matching score wasbetter than the tab-matching score by 5%. The meanmatching scores for the variables of gender, tabs, disks,and order of porcelain design matching (tab or disk first)are presented in Table I. Female observers demonstratedvirtually the same matching ability among themselveswith the disks and tabs, matching the tabs only 1% moreaccurately than the disks. Within the male observergroup, disks were matched 8% more accurately thantabs. In a comparison by gender, men matched the disks3% more frequently than women, whereas womenmatched the tabs with approximately 6% more accuracy.However, there was no significant difference betweenthe genders in terms of matching ability (P".21).

The comparison of disk selection before tab selectionyielded equivalent levels of shade-matching accuracy ineach exercise. However, when tab selections were thefirst exercise, improvement was evident on the second

Table II. Error scores for color-deficient vision observers

Isochromaticplates (%)

Saturated 15 HueTest (%)

Disk matching(%)

Tab matching(%)

53.3 0 25.0 37.553.3 2 37.5 12.560.0 78 0 62.566.6 68 50.0 50.0

Zero represents a perfect score in all 4 categories. Two of four observers scored“normally” on the 15-hue test, and one observer matched all 8 disks perfectlydespite the fact that the isochromatic plates indicated color-defective vision.

Table I. Percent correct matches overall, by gender, andby specimen shape

Correct matches (%)

Overall Female (n ! 25) Male (n ! 48) Tab/disk Disk/tab

Disks 78.4 76.5 79.4 79.8 77.6Tabs 73.6 77.5 71.6 67.3 77.1P value .119 .826 .054 .010 .892

Tab/disk " Tab matching followed by disk matching; disk/tab " disk matchingfollowed by tab matching.

BARRETT ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

DECEMBER 2002 593

shape-matching test (P".01): the disks were matched12.5% more accurately than the tabs. No learning wasdemonstrated when the tabs were selected after thedisks. This order yielded virtually the same number ofmatches with disks and tabs (77.6% and 77.1%, respec-tively).

The results of the 2 standardized color tests indicatedthat there were 4 color-blind men among the observers.None of the dental observers were rejected as a result ofstandardized test scores, because these persons mightrepresent the diversity of clinicians’ ability. Of the 4color-blind observers, 2 were not aware of their colorvision deficiencies before the screening. Table II lists thescores for the color-deficient observers. Note that 1 col-or-deficient observer was able to match all of the disksaccurately, demonstrating that one can be color vision-deficient yet still be able to discriminate differences well.The same observer, however, was able to match only37.5% of the shade tabs correctly.

Tables III and IV combine the selections for all ob-servers. The highest percentages recorded represent cor-rect matches. Table III shows 3 matches chosen at orabove 20%; these represent errors in color/shade per-ception. Disk A3 was paired as a match with A2 by 29%of the observers (Fig. 1), C2 with B3 by 20% of theobservers (Fig. 2), and D2 with C1 by 34% of the ob-servers (Fig. 3). There were 15 other incorrect pairingswith frequencies below 20%.

As shown in Table IV, the correct pairings for the tabsgenerally were associated with the highest percentages

shown, although the matching values were not signifi-cantly different from the disk values. The correctmatches ranged from 43% to 99%. Shade tabs A3 and B2had the highest percentages of correct matches (92% and99%, respectively). Shade tab A1 was inappropriatelypaired with C2 by 52% of the observers (Fig. 4), whereasonly 43% assigned shade tab A1 to A1. Only 2 additionaltab shades were incorrectly paired by more than 20% ofthe observers: 22% chose shade tab A2 as a match withA1 (Fig. 5) and 22% selected shade tab B3 as a matchwith B2 (Fig. 6). There were 30 other incorrect pairingsbelow 20%, 24 of which were below 5% (Table IV). Agreater number of correct disk matches occurred forshades A1, A2, and B3, whereas a greater number ofcorrect tab matches occurred for shades A3, B2, C1, andD2.

The most frequently matched shades (Table V) dif-fered for the disks and tabs, with the exception of shadeC2, which was paired by 77% of the observers in bothshapes. Shade A1, which is the highest in value, wasaccurately matched 100% of the time for the disk shape.However, the tooth-shaped shade tab A1 was pairedaccurately by only 43% of the observers.

None of the 73 dental observers mismatched morethan 5 of 8 possible selections for either shape. In thedisk selection, 9.6% of the observers had 50% error rates.The female observers mismatched 4.1% of the disks, andthe male observers mismatched 5.5%. The overall taberror rates were 8.2%, all within the male observer pop-ulation; however, this did not constitute a significant

Table IV. Tab-matching selections of all 73 observers

A1 A2 A3 B2 B3 C1 C2 D2 Total

A1 42.5% 1.4% 1.4% 1.4% 1.4% 52.1% 100%A2 21.9% 72.6% 1.4% 1.4% 1.4% 1.4% 100%A3 91.8% 1.4% 2.7% 2.7% 1.4% 100%B2 1.4% 98.6% 100%B3 11% 21.9% 63% 1.4% 2.7% 100%C1 13.7% 4.1% 1.4% 76.7% 4.1% 100%C2 1.4% 6.8% 2.7% 2.7% 1.4% 76.7% 8.2% 100%D2 2.7% 11% 13.7% 1.4% 69.9% 100%

Correct selection percentages are underlined.

Table III. Disk-matching selections of all 73 observers

A1 A2 A3 B2 B3 C1 C2 D2 Total

A1 100% 100%A2 95.9% 1.4% 2.7% 100%A3 28.8% 61.6% 8.2% 1.4% 100%B2 16.4% 1.4% 78.1% 4.1% 100%B3 16.4% 80.8% 2.7% 100%C1 9.6% 12.3% 74% 4.1% 100%C2 2.7% 20.5% 76.8% 100%D2 1.4% 1.4% 34.2% 63% 100%

Correct selection percentages are underlined.

THE JOURNAL OF PROSTHETIC DENTISTRY BARRETT ET AL

594 VOLUME 88 NUMBER 6

difference in shade-matching ability by gender. The menmismatched 10.9% of the tab selections, whereas thewomen mismatched 2.7%.

The rates of mismatches of only 1 or 2 selections outof 8 possibilities were 56.1% for disks and 42.5% overall.

Of the female observers, 20.5% missed 1 or 2 disks and16.4% missed 1 or 2 tabs; for male observers, the samemeasurements were 35.5% and 25.9%, respectively.

The overall percentage of the observer populationwith no errors in shade matching was 20.5% for disks

Fig. 1. Disk shade A3 selected as match with disk shade A2by 29% of 73 dental observers.

Fig. 2. Disk shade C2 selected as match with disk shade B3by 20% of 73 dental observers.

Fig. 3. Disk shade C1 selected as match with disk shade D2by 34% of 73 dental observers.

Fig. 4. Shade tab A1 selected as match with shade tab C2 by52% of 73 dental observers.

Fig. 5. Shade tab A2 selected as match with shade tab A1 by22% of 73 dental observers.

Fig. 6. Shade tab B2 selected as match with shade tab B3 by22% of 73 dental observers.

BARRETT ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

DECEMBER 2002 595

and 19.1% for tabs. The female zero-error results were5% for disks and 6.8% for tabs. The male zero-errorresults were 15% for disks and 12.3% for tabs.

DISCUSSIONNo significant difference in shade-matching accuracy

(P%.05) was found between the disk and tab specimens.Nevertheless, the results indicated that the order inwhich specimen designs were matched (disks or tabsfirst) influenced shade-matching ability for the subse-quent exercise. Farnsworth14 found that improvementgenerally was demonstrated during a second test as aresult of familiarity with the materials. In the presentinvestigation, however, when tab matching followeddisk matching, the rate of accuracy was virtually the samefor the 2 shapes. Given that the flat disks were matchedwith greater accuracy (12.5%) than the shade tabs, someslight complexity in matching with a tooth-like speci-men may be indicated.

Value (brightness) has been considered the mosthighly recognizable characteristic in dental shadematching.15 Shade tab A1, the second highest value inthe Vita set and the highest value in the present study,was matched by 100% of the population when the shadewas in disk form but by only 42% when the shade was intab form. These results indicate that value may not nec-essarily be the most important (highly recognized) fac-tor in matching. In fact, A1 was matched to C2 by 52%of the observers. These 2 shades were separated by 5other shades in the viewing arrangement. Two othermismatched pairs were shade tab A2 with A1 and B3with B2. Both pairs were misperceived by 22% of theobservers. Although they constituted mismatches, bothcombinations were within the same shade group yetdiffered in value and saturation. As stated previously, thefrequency of assignments differed for the disks and tabs,demonstrating the irregularities or variability of shadematching. The disks were homogeneous with a glossyfinish, whereas the tabs, like natural teeth, had surfaceirregularities that affected the gloss (light reflection on

the surface). The varying zones of translucency in thetabs also may account for some of the misperceptions.

As reported in the dental literature, variability in shadematching may be affected by different factors such as light-ing, surround, fatigue, and other distractions, as well as bythe lack of color perception development. This study wasexecuted under almost ideal conditions of 1 unchangingsurrounding environment and 1 single, consistent illumi-nant. Even so, there were still notable inconsistencies inshade recognition and matching. The results for this dentalstudy population dispel the popular conception thatwomen are better than men at color matching. One mayinfer that the dental assistant (female or male) could havethe same, though not necessarily better, shade-matchingability as the dentist with normal color vision. Anusavice16

reported no difference in matching ability relative to gen-der, age, or clinical experience with 116 observers. Onemay conclude that the dentally educated population ismore discriminating visually than the population at large.

Objective knowledge of one’s color vision is an im-portant factor in judging shade-matching ability. It isimportant to note that one color-blind observer was ableto match all 8 disks perfectly, demonstrating the signif-icance of discrimination ability. The clinical ramifica-tions of this study are easily seen when one considersthat 76% of the observers missed 2 or fewer disk selec-tions, indicating that 24% of a potential patient popula-tion would receive mismatched restorations. Two orfewer mismatches of tooth-shaped tabs were recordedfor 61% of the observers. This also may be construed toindicate that almost 40% of the patients seen by thesedentists would receive mismatched restorations and that40% of the dentists might not perceive the mismatches inthose restorations.

Subjective clinical shade matching of individual teethand attendance at limited lectures are not sufficientlearning tools for esthetic development, as substantiatedin the dental literature. Color recognition and shade-matching ability can be learned. As demonstrated in thisstudy, the visual acuity required for matching polychro-matic shades in dentistry is the same as that required formatching monochromatic porcelains. These skills maybe improved through practical, organized instruction.Further development and delivery of instructive, educa-tional color acuity programs in dental education and incontinuing education may improve the individual clini-cian’s color assessment. Practical, repetitive teaching ex-ercises would benefit both the dentist and the patient inrestorative and esthetic dentistry. As the public becomesmore aware and more educated with regard to esthetics,so must the dentist.

CONCLUSIONSWithin the limitations of this study, the following

conclusions were drawn:

Table V. Percent matches for each shade and shape

Shade

Matches by shade and shape (%)

Disks Tabs

A1 100.0* 42.5A2 95.9* 72.6A3 61.6 91.8*B2 78.1 98.6*B3 80.8* 63.0C1 74.0 76.7*C2 76.7* 76.7*D2 63.0 69.9

*Most frequently matched shades.

THE JOURNAL OF PROSTHETIC DENTISTRY BARRETT ET AL

596 VOLUME 88 NUMBER 6

1. No significant significance was found in matchingaccuracy for polychromatic tooth-shaped tabs and ho-mogeneous porcelain disks. One design was not moredifficult to match than the other.

2. The order in which matching was performed (disksor tabs first) influenced shade-matching ability. The resultssuggest that proficiency in shade matching can be learnedand developed through multiple objective exercises, asdemonstrated when tabs were matched before disks.

3. Gender had no significant relevance to dentalshade-matching ability.

4. Of the 4 color-blind observers, 2 were not awareof their color vision deficiencies before the standardizedcolor vision screening exercises. All practitioners maybenefit from an objective knowledge of their own colorvision status.

REFERENCES1. Clark EB. An analysis of tooth color. J Am Dent Assoc 1931;18:2093-103.2. Sproull RC. Color order systems and their application in dentistry. In:

Yamada HN, editor. Dental porcelain: the state of the art-1977. LosAngeles: University of Southern California; 1977. p. 317-21.

3. Moser JB, Wozniak WT, Naleway CA, Ayer WA. Color vision in dentistry:a survey. J Am Dent Assoc 1985;110:509-10.

4. Rawlinson A. A simple eyesight screening programme for dental under-graduates: results after 7 years. Aust Dent J 1993;38:394-9.

5. O’Keefe K, Strickler ER, Kerrin HK. Color and shade matching: the weaklink in esthetic dentistry. Compendium 1990;11:116, 118-20.

6. Wasson W, Schuman N. Color vision and dentistry. Quintessence Int1992;23:349-53.

7. van der Burgt TP, ten Bosch JJ, Borsboom PC, Plasschaert AJ. A newmethod for matching tooth colors with color standards. J Dent Res 1985;64:837-41.

8. CIE Colorimetry Committee. Technical notes. Working program on colordifferences. J Opt Soc Am 1974;64:896-7.

9. Bolt RA, ten Bosch JJ, Coops JC. Influence of window size in small-windowcolor measurement, particularly of teeth. Phys Med Biol 1994;39:1133-42.

10. Clarke FJ. Measurement of color of human teeth. In: McLean JW, editor.Dental ceramics. Proceedings of the First International Symposium onCeramics. Chicago: Quintessence; 1983. p. 441-88.

11. Goodkind RJ, Schwabacher WB. Use of a fiber-optic colorimeter for in vivocolor measurements of 2830 anterior teeth. J Prosthet Dent 1987;58:535-42.

12. Seghi RR, Hewlett ER, Kim J. Visual and instrumental colorimetric assess-ments of small color differences on translucent dental porcelain. J DentRes 1989;68:1760-4.

13. Okubo SR, Kanawati A, Richards MW, Childress S. Evaluation of visualand instrument shade matching. J Prosthet Dent 1998;80:642-8.

14. Farnsworth D. The Famsworth-Munsell 100-hue and dichotomous tests forcolor vision. J Opt Soc Am 1943;33:586-78.

15. Shillingburg H, Hobo S, Whitsett LD, Jacobi R, Brackett SE. Fundamentalsof fixed prosthodontics. 3rd ed. Chicago: Quintessence; 1997. p. 437.

16. Anusavice KJ, editor. Phillips’ science of dental materials. 10th ed. Phil-adelphia: Saunders; 1996. p. 42.

Reprint requests to:A.A. BARRETT

DEPARTMENT OF DENTAL BIOMATERIALS

PO BOX 100446COLLEGE OF DENTISTRY

UNIVERSITY OF FLORIDA

GAINESVILLE, FL 32610-0446E-MAIL: abarrett@dental.ufl.edu

Copyright © 2002 by The Editorial Council of The Journal of ProstheticDentistry.

0022-3913/2002/$35.00 & 0 10/1/129892

doi:10.1067/mpr.2002.129892

Noteworthy Abstractsof theCurrent Literature

Materials design of ceramic-based layer structures for crownsLawn BR, Deng Y, Lloyd IK, Janal MN, Rekow ED,Thompson VP. J Dent Res 2002;81:433-8.

Purpose. This study examined the hypothesis that critical loads for radial cracking in crown-likelayers vary with the square layer thickness (at ceramic thickness less than 1 mm).Material and Methods. Crown-like ceramic layers were bonded to simulated dentin substratesusing an epoxy adhesive. The 4 ceramics types used were: a porcelain (Mark II; Vita Zahnfabrik,Bad Sackingen, Germany); an infiltrated alumina (In-Ceram; Vita Zahnfabrik); a zirconia (Prozyr;Norton Advanced Ceramics, Colorado Springs, Colo.); and a glass-ceramic (Dicor; DentsplyInternational, York, Penn.). The bonded ceramic thicknesses ranged from 100 !m to 6 mm.Hertzian tests were completed on the layer specimens. Radial crack initiation and evolution werevideo recorded in situ during loading. Critical loads were recorded. Regression analyses werecompleted on the data and correlation coefficients were ascertained.Results. An increase in resistance to radial cracking was demonstrated for zirconia relative toalumina and also for alumina relative to porcelain.Conclusion. This investigation provided failure predictions in ceramic/substrate layers. Four typesof ceramic materials were also ranked in terms of predicted clinical performance. 31 references.—DL Dixon

BARRETT ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

DECEMBER 2002 597

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 276–281

avai lab le at www.sc iencedi rec t .com

journa l homepage: www. int l .e lsev ierhea l th .com/ journa ls /dema

Visual and instrumental agreement in dental shadeselection: Three distinct observer populations and shadematching protocols

Alvaro Della Bonaa,!, Allyson A. Barrettb, Vinicius Rosac, Caroline Pinzettaa

a Department of Restorative Dentistry, University of Passo Fundo, RS, Brazilb Department of Dental Biomaterials, University of Florida, Gainesville, FL, USAc University of São Paulo, São Paulo, SP, Brazil

a r t i c l e i n f o

Article history:Received 11 July 2008Received in revised form18 September 2008Accepted 27 September 2008

Keywords:Visual shade matchInstrumental shade identificationShade guideColorLighting

a b s t r a c t

Objectives. This study tested the hypothesis that the agreement between observer visualdental shade matches and instrumental shade identification is higher using the Vita 3D-Master® (3D) shade guide than the Vita classical (VC) shade guide.Methods. Three populations selected shade matches: non-dental observers (GP) matchedshade tabs-to-tabs and dental students (DS) and dentists (DD) matched an in vivo naturalright upper central incisor (RUCI). All observers (n = 600) used both shade guides (3D and VC)in two lighting conditions, cool white fluorescent lighting (CWF) and natural sunlight (NSL).The shade tabs and natural teeth were identified using an intra-oral spectrophotometer (VitaEasyshadeTM) to determine the instrumental agreement with the visual shade selection. Thepercent visual–instrumental shade agreement (PVIA) was analyzed statistically considering:observer population, shade guide set, and lighting condition.Results. A “substantial” intra-examiner agreement (k = 0.76) was observed. The PVIA rangedfrom 12% (DS) for the 3D-NSL condition to 42% (DD) with the VC-CWF condition, which alsoresulted in the highest PVIA for GP (38.5%) and DS (35%). Results indicated that the GP withneither dental knowledge nor shade guide experience had a significantly higher PVIA usingthe VC rather than the 3D shade guide. Dentists demonstrated the highest PVIA (42%) forboth lighting conditions and shade guides, thereby rejecting the study hypothesis.Significance. A significantly higher visual–instrumental shade agreement was demonstratedby the clinically experienced dentists (DD), regardless of shade guides and lighting condi-tions.

© 2008 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

1. Introduction

Esthetic excellence is largely an art with primarily subjectiveinterpretation and not enough has been done to effectivelyanalyze and formulate it [1].

! Corresponding author at: Department of Restorative Dentistry, University of Passo Fundo, Campus I - BR 285, PO Box 611, Passo Fundo,RS 99001-970, Brazil. Tel.: +55 54 3316 8402; fax: +55 54 3316 8403.

E-mail address: dbona@upf.br (A. Della Bona).

Included within restorative and esthetic dentistry is color,in the practice of shade matching. Today’s color science prin-ciples still originate with Newton in the 1600s and are stillbased on Munsell’s basic three-dimensional notation theoryof the early 1900s [2].

0109-5641/$ – see front matter © 2008 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.dental.2008.09.006

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 276–281 277

It is the subjectivity inherent in the shade matching processthat people try to overcome [3]. Yet, in dental shade match-ing, the human eye is still the most popular clinical approach.A person with color normal vision can recall approximately300 different colors and is able to discriminate 5–10 milliondifferent colors in side-by-side comparisons [4]. Even so, stud-ies have reported that up to 80% of the patients express theirdissatisfaction with perceptible shade differences [5]. It hasbeen affirmed that in dental shade matching, the eye is thefinest null detector, nonetheless, observers must be trained tooptimize their color perception [6]. Shade matching is gener-ally associated with homogeneously colored objects however,teeth vary in color and translucency. Consequently, they arethought to be more difficult to shade match. A study using 73dentally trained observers reported no significant differencebetween matching accuracy for the homogeneous porcelaindiscs (78.4%) and the inhomogeneous shade tabs (73.6%) [7].

Given that the same color can be perceived differentlyamong observers [8], it is feasible that instrumental shadeidentification may remove a certain subjectivity that arisesfrom individual color perception. Addressing the issue of colorvision disparities, Moser et al. [9] screened the red-green colorvision of 670 participants at an ADA meeting where nearly 13%of dental professionals demonstrated irregular color vision,while 2.8% exhibited severe color vision irregularities.

However, it has been demonstrated that instruments alsohave limitations. One such study reported 50% instrumen-tal accuracy and 48% visual accuracy in matching designatedcommercial tabs, with no statistically significant differencebetween the two methods [10]. Another study demonstratedinconsistent interexaminer reliability for instrumental shadeidentification among four clinicians with different levels ofcolor instrument training; although, the instrumental resultsfor canine were more reproducible compared to those ofcentral incisors [11]. Nonetheless, many dental studies havereported that color-measuring instrumentation has facilitatedand supported the clinician’s shade selection to match thesurrounding dentition [3,11–16].

One such instrument is an “intra-oral spectrophotometer”(EasyshadeTM, Vita Zahnfabrik, Bad Säckingen, Germany) withspecific modes to identify reference shades from two com-mercial dental shade guides: the Vita 3D-Master (3D) (VitaZahnfabrik, Bad Säckingen, Germany) and the Vitapan Clas-sical (VC) (Vita Zahnfabrik, Bad Säckingen, Germany). Thiscombination of shade guides, instrumental development andcalibration capability presents a very orderly, integrated sys-tem for testing. The Easyshade is reported to convert thereceived light from the targeted object into tristimulus param-eters based on a D65 illuminant, while the unit itself uses a20 W halogen bulb. The color-measurement area includes twoseparate measuring apertures, a 1-mm and a 3-mm diameter.The illuminating diameter is 5 mm.

Dozic et al. [17] reported that the intra-oral spectropho-tometer, Easyshade, was the most reliable instrument in bothin vitro and in vivo circumstances. However, the relationship ofhuman visual observations to instrumental shade identifica-tions for dental applications is still lacking. There are very fewreports with a substantial number of visual observations indirect comparison with instrumental evaluations. Thus, basedon limited observer matches in conjunction with instrumen-

tal evaluations, a wide range and overlap of shade matchesand mismatches have been reported [6,18–20].

While color instrumentation and shade matching proce-dures have been widely addressed in dental literature, themost popularly used shade guides have not changed muchthrough the last 50 years. The Vita Toothguide 3D-Master(3D) was developed with a systematic arrangement for a widerange of natural dentition shades [21]. The 3D shade guide isarranged in five discernible value levels with multiple chromalevels, as differentiated from the traditional Vitapan Classical(VC) grouping primarily by hue.

As is known, lighting is fundamental to color perceptionand significantly influences the quality of shade matching.Studies have reported that shade matching performed underdifferent lighting is not necessarily consistent and further-more, it is difficult to determine the effects of particularlighting on shade perception [22–24]. Consequently, controlledand standardized illumination is often suggested for observerand instrumental dental shade evaluation. Considering thatthe instrument used in the present study is based on a day-light (D65) illuminant and that most clinical shade matchingoccurs under cool white fluorescent lighting, both of thoseconditions were employed in this study. While these are notnecessarily the most accurate illumination sources for shadematching, they are among the more popularly used. Whenpossible, consideration should be given to such variables asthe color rendering index (CRI).

This study analyzed the percent agreement between thehuman observer visual shade match selections and the cor-responding instrumental shade identifications, that is the“percent of visual–instrumental agreement” (PVIA) amongthree populations with varying levels of dental experience,testing the hypothesis that the agreement of visual shadeselections and the instrumental shade identification is higherwith use of the 3D shade guide than with the VC shade guide.

2. Materials and methods

The study protocol was presented to and approved by the localEthics in Research Committee prior to the experiment. Allthe individuals involved were invited to participate in dentalshade matching exercises on a voluntary basis. Prior to shadematching participation, each individual was screened for colorvision recognition ability by correctly identifying selectedplates from the Ishihara Color Vision Test. The voluntaryobservers fell into one of three distinct categories predicatedon their dental knowledge and dental shade experience: (1)general population (GP) with no dental shade experience, (2)first year dental students (DS) with no clinical shade experi-ence, and (3) dentists (DD) with a range of years experience.The gender and age of the participants were recorded.

Shade tabs from two commercial dental shade guides (VCand 3D) were used for all of the experimental shade match-ing exercises. The shade matching protocols differed to someextent in their implementation, appropriate to the expe-rience and classification of each observer population. TheGP participants matched shade tab-to-tabs. The DS and DDgroups conducted intra-oral shade matches. Only the DD hadpatient subjects. For intra-oral shade matching, subjects and

278 d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 276–281

patients were upright with the mouth at the observer’s eyelevel.

External visual influences such as lipstick were removedand a neutral gray patient bib obscured clothing colors. Upperand lower teeth were apart and the tongue retracted. Theshade tab was positioned in the same plane as the tooth (rightupper central incisor—RUCI) to be matched by DS and DD [25].

Following each observer’s visual shade tab selection thecorresponding shade tab or natural dentition was identifiedinstrumentally using an intra-oral spectrophotometer (VitaEasyshade®, serial no. 405643). The stainless steel probe tip ofthe instrument was protected with a specified polyurethaneshield (Vita Zahnfabrik) for contamination control. Calibra-tion was performed throughout the study. The ‘instrument usemode’ was set on the appropriate identification mode, that is,to “shade tab” for identification of tab shade and to “toothsingle” to identify the intra-oral tooth shade. In addition, theinstrument was set to yield shades in the commercial nomen-clature for both of the shade guides (VC and 3D) that were usedthroughout this study.

Shade matching selections were conducted in two dif-ferent lighting environments: ‘out-of-doors’, natural sunlight(NSL) and ‘in-doors’ under cool white fluorescent (CWF) light-ing. Throughout the course of the study, the natural sunlightexercises were conducted during the same daylight hours,in keeping with dental clinic hours. As commercial shadetabs were used, the shade nomenclature on the tab han-dle was masked to prevent influencing observers’ selections.The experimental adaptations particular to each population’sexperience are clarified below.

2.1. General population (GP)

Participants were enlisted by inviting random passersby in thevicinity of the dental school to engage in dental shade match-ing exercises. No explanation of color science principles or ofclinical shade matching procedures was given to the partici-pants. GP (n = 200) were instructed individually to focus on themiddle third of the shade tab when selecting the shade tab-to-tab matches. The middle third focus corresponded to thearea targeted with the instrument. Each observer was askedto take one shade tab from a box containing eight shades(Table 2) and to select the matching tab from each of twoshade guide sets (VC and 3D). All identifying shade nomen-clature was masked to prevent influencing shade selections.After matching in each of the two lighting conditions (NSL andCWF), each GP observer shade selection was identified instru-mentally and recorded as a visual–instrumental “agreementor non-agreement”.

2.2. First year dental students (DS)

DS participants (n = 200) were divided into pairs to conductthe shade matching. Each DS selected, in vivo, a shade tabmatch for a natural right upper central incisor (RUCI) ontheir partner. All participating DS had unobscured, natu-ral dentition (RUCI), meaning there were no brackets orrestorations to interfere with the visual shade assessmentor instrumental shade identification. DS were instructedto focus on the middle third of the tooth and the shade

tab during matching, which corresponded with the place-ment of the instrumental probe. The visual shade selectionsmade in CWF and NSL were identified as visual–instrumental“agreements or non-agreements” in accordance with theinstrumental intra-oral shade identification of the designatedRUCI.

2.3. Dentists (DD)

DD participants (n = 200) followed same protocol as the DSand selected, in vivo, a shade match for a patient’s naturalRUCI. As with the other observer populations, the DD wereinstructed to focus on the middle third of the RUCI beingmatched. The patient’s RUCI was clear of restorations or oralappliances that might interfere with the visual and instru-mental shade identification. After visual shade selection forthe RUCI in CWF and NSL, intra-oral instrumental identifi-cation was conducted and recorded as visual–instrumental“agreement or non-agreement”.

All the instrumental shade identifications were made threetimes for each shade tab and each natural tooth after eachobserver’s shade selection. The shielded instrumental probewas consistently placed in the middle third of the tab or toothto correspond to the observer viewing instructions. Any dis-crepancy in the instrumental multi-readings was resolved byperforming an additional set of readings per the manufac-turer’s recommendation. One single consensus was recordedfor each visual–instrumental determination.

The intra-examiner Kappa coefficient (k) was calculatedas previously described [26]. Each observer shade selectionand instrumental shade identification was recorded strictly asa visual–instrumental “agreement or non-agreement” basedon the instrumental shade identification. The data was com-piled and analyzed statistically using Pearson’s chi square test(˛ = 0.05) considering the observer population, shade guide set,lighting condition and visual–instrumental shade agreementor non-agreement.

3. Results

A “substantial” intra-examiner agreement (k = 0.76) wasobserved [26]. This study involved a total of 600 observers whoranged from 19 to 50 years of age. There were 299 females and301 males with distinctly different levels of knowledge andexperience for dental shade matching (Table 1). Each observermade four shade match selections, using two shade guides(VC and 3D) in two lighting conditions (NSL and CWF) for atotal of 2400 observer shade match selections. The percentvisual–instrumental agreement (PVIA) for each shade guide –illuminant condition per population – was calculated (Table 1).

Approximately 7% of the individuals that volunteered toparticipate in this study were excluded prior to shade match-ing because they were unable to identify all the color testplates indicating possible color vision confusion. By gender,the exclusion was 10.4% males (35 of 336) and 2.3% females(7 of 306). Per population, the ineligible volunteers were: GP:12 (6%), 2 females and 10 males; DS: 14 (7%), 2 females and 12males; and DD: 16 (8%), 3 females and 13 males. These resultsare in agreement with previous reports [9,13,16].

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 276–281 279

Table 1 – The number of observers (n) by gender (F, female; M, male) and the PVIA (%) per shade guide, lighting conditionand observer population, with statistical groupings.

n F - M VC-NSL (%) 3D-NSL (%) VC-CWF (%) 3D-CWF (%)

GP 105 - 95 32.0AB 20.5B 38.5A 23.5BDS 103 - 97 29.5B 12.0C 35.0A 23.0BDD 91 - 109 41.0A 41.5A 42.0A 40.5A

Different letters within each column represent statistical groupings (p < 0.05).

The shade tabs used in this study were identified instru-mentally confirming the shade tab identification prior to use.The instrumental identification indicated that there wereeight equivalent shades in the two different shade guide sets(VC and 3D) (Table 2).

Using the specific shade guide mode for the instrument,shade tab identification was consistently in agreement withthe tab nomenclature. This was verified by unmasking the tabnomenclature after each GP observer completed their shadeselections.

The most consistent matching condition, with no statisti-cal difference among the three observer groups, was for theVC shade guide under CWF lighting (Table 1).

Combining the matching results for both shade guides,the GP visual–instrumental agreement (PVIA) was significantlyhigher when observers selected shades under CWF lighting(31%), rather than under NSL (26.2%) (p < 0.05). When the GPshade tab selections under both lighting conditions were com-bined, the PVIA was significantly higher using the VC shadeguide (35.2%), than for the 3D shade guide (22.0%) (p = 0.0001).

Combining the results of the two lighting conditions, theDS demonstrated significantly greater PVIA using VC (32.2%)than using 3D (17.5%) (p = 0.0001). The in vivo RUCI PVIA wassignificantly higher under the CWF lighting (29%) than underthe NSL (20.7%) (p = 0.0067), when the results of the two shadeguides were combined.

Considering the VC guide selections made under NSL, thedentists (DD) PVIA (41%) was significantly higher than thatfor the DS (29.5%) (p = 0.039). Although, there was no signifi-cant difference among the three groups when using the VCguide under CWF lighting (p = 0.3). However, when using the3D guide in NSL the DD PVIA (41.5%) was significantly greaterthan that of the other two populations (p = 0.0001). The GPtab-to-tab PVIA was significantly higher than the DS shadetab-to-natural dentition with the 3D-NSL condition (p = 0.021),

Table 2 – Instrumental identification for same shades intwo shade guide sets.

Eight equivalent shades

VC shades 3D shades

A1 1M2A2 2M2A4 4L2.5B1 1M1B4 3M3C1 2L1.5C3 4L1.5C4 5M2

but these populations (GP and DS) demonstrated similar PVIAwith the other conditions. Yet, the DD demonstrated a signif-icantly greater PVIA using the 3D in both lighting conditionsthan the other two population groups (p = 0.0001). There wasno statistical difference between either the shade guide or thelighting conditions influence on PVIA for the DD population(p > 0.05).

When the results for the two shade guides (VC and 3D)under NSL (1200 observations) were grouped and analyzed,there was no significant difference for the PVIA between thetwo shade guides (p = 0.71). Nor was there any statistical differ-ence (p = 0.23) between the lighting conditions (NSL and CWF)when all the results for visual–instrumental agreement werecompared among the total 2400 visual observations. However,in the more consistent cool white fluorescent lighting (1200observations) the PVIA for the VC guide (38.5%) was signifi-cantly higher than for the 3D shade guide (29%) (p = 0.0001).

4. Discussion

As previously stated, the percentages reported in this studyreflect the percent agreement between the visual shade selec-tions and the instrumental shade identifications (PVIA).

Previous reports have indicated that spectrophotometersare the most reliable standard for color matching studies[17,25]. One study concluded that instrumental dental shadeanalysis was as much as 33% more accurate than visual shadeevaluation [13], however another study reported no statisticaldifference [10]. The authors of the present study think thatwhile shade matching instrumentation has the potential toremove some subjectivity, it is important to determine to whatextent instrumental shade identification agrees with visualobservations of calibrated observers.

The testing hypothesis throughout this study was that thevisual–instrumental shade identification agreement (PVIA)would be higher using the 3D shade guide. This hypothesiswas based, in part, on the increased shade range selectionof 26 3D shades rather than the familiar 16 VC shades andalso on the basis that the 3D shade guide design presents anew viewing arrangement for value and chroma that is asso-ciated with instructions of use. However, it must be noted thatthe number of samples in the shade guide or their arrange-ment should have no effect on the instrumental functioningfor shade identification.

The results of this study demonstrated that a general pop-ulation (GP) with neither dental knowledge nor shade guideexperience had a higher PVIA when they used the VC shadeguide, thus rejecting the hypothesis for this population. Thesame held true for the first year dental students (DS) in vivo

280 d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 276–281

matching selections, which demonstrated higher PVIA whenthey used the VC shade guide than the 3D. These two pop-ulations (GP and DS) had neither previous experience norknowledge of the 3D shade guide instructions of use.

Dentists (DD) demonstrated the highest PVIA (42%) and rel-ative consistency under both lighting conditions, which mayindicate that the dentists were not influenced by a particu-lar shade guide set, rejecting the study hypothesis for thispopulation. A small number of the dentists (14%) reported pre-vious awareness of the 3D shade guide. That may have been acontributing factor to the group demonstrating a statisticallysignificant greater PVIA than the other two populations withthe 3D shade guide. Also, it may be considered that familiaritywith the shade matching process among dentists played a roleamong the populations.

Another consideration for visual–instrumental agreementis that the human eye sees the heterogeneous aspects ofeach tab or tooth, the variations in shade and translucency.An instrument amalgamates these characteristics yieldingone, homogeneous shade. The instrumental shade identifica-tion is affected by the amount of light that is reflected backinto the instrument from the surface being targeted. Conse-quently, positioning of the probe or mouthpiece is reportedto be critical to the repeatability of the instrumental read-ing. For that reason, one trained operator (k = 0.76) performedthe instrumental identification throughout the presentstudy.

The better standard illuminant for color viewing is gener-ally thought to be the natural daylight, however it changes overtime, especially given the significant number of observer hoursin this study. While two lighting environments were used, thecool white fluorescent lighting may have provided a more con-sistent illumination than the natural sunlight. This may havehad some effect on the visual selections, and subsequentlyon the study results. This observation is in accordance withliterature indicating that the consistency of artificial lightingmay contribute to better shade-matching results than naturaldaylight [15,27].

Simply stated, the color differences from color-measuringdevices are based on the light emitted and reflected back fromthe targeted object. As noted by Ishikawa-Nagai et al. [16],samples need to be well illuminated to avoid measurementerrors. The instrument in this study has an illuminating areaof 5 mm, incorporating each of the two smaller shade identifi-cation aperture areas. However, an instrument is not affectedby most surrounding visual influences as is the human eye.At present, shade matching usually employs different illumi-nants such as those used in this study: environmental lighting(CWF "4000 K and NSL "5000–6500 K), instrumental lighting(halogen, 3350 K) and the instrumental color calculations (D65,6500 K). Therefore, it would be advantageous to use the samestandardized lighting for both the instrumental illuminantand for visual perception shade-matching studies if the goalis comparing results from different studies. This study did usenatural daylight, which is purported to correspond to the D65

used in instrumental calculations. The CWF used in this studyis closer to the instrumental illuminant than the NSL. In addi-tion, it should be noted that these study results demonstratedthe consistent and reliable calibration of the Easyshade bythe correct shade recognition of the tabs. This concurs with

previous studies [17,26], and it was the reason for using thisinstrument as the gold standard in the present study.

This research involved a significant body of observers anda correspondingly large number of visual dental shade selec-tions using the same dental shade tabs under two differentlighting conditions as recommended in the ASTM D-1729-89[28]. Although the emphasis of this study was on the agree-ment for the observer visual and the instrumental shadeidentification, it is an harmonious appearance that is mostessential to successful dental restorations.

Although a significantly higher visual–instrumental shadeagreement was demonstrated by the clinically experienceddentists (DD), it should be possible to achieve a higher per-centage of successful matches than the approximately 50%reported in the dental literature [10,14,16], which are in accor-dance with the results of the present study. Therefore, theresults of this study suggest that shade training and/or dentalexperience are an important component in shade matching.As stated by Sproull, concentrated effort on shade training cancontribute to improve matching results [6].

Further study with an experienced population that is famil-iar with the shade guide systems may yield different andhopefully higher agreement results.

Acknowledgment

The authors acknowledge the assistance of the statistician, Dr.Dileta Cecchetti.

r e f e r e n c e s

[1] Preston J. The elements of esthetics—application of colorscience. In: McLean JW, editor. Dental ceramics: proceedingsof the first international symposium on ceramics. Chicago:Quintessence; 1983. p. 491–520.

[2] Nickerson D. History of the Munsell color system, vol. 7(5).New York: Color Engineering, Chromatic CommunicationsInc.; 1969.

[3] Yap AU, Sim CP, Loh WL, Teo JH. Human-eye versuscomputerized color matching. Oper Dent 1999;24:358–63.

[4] Hunter RS, Harold RW. The measurement of appearance.New Jersey: John Wiley and Sons; 1987.

[5] Kawaragi C, Ishikawa S, Miyoshi F, Furukawa K, Ishibashi K.Evaluations by dentists and patients concerning the color ofporcelain fused-to-metal restorations. Dent J Iwate MedUniv 1990;15:9–17.

[6] Sproull RC. Color order systems and their application indentistry. In: Yamada NH, editor. Dental porcelain: the stateof the art—1977. Los Angeles: University of SouthernCalifornia; 1977. p. 317–21.

[7] Barrett AA, Grimaudo NJ, Anusavice KJ, Yang MC. Influenceof tab and disk design on shade matching of dentalporcelain. J Prosthet Dent 2002;88:591–7.

[8] Land EH. The retinex theory of color vision. Sci Am1977;237:108–28.

[9] Moser JB, Wozniak WT, Naleway CA, Ayer WA. Color visionin dentistry: a survey. J Am Dent Assoc 1985;110:509–10.

[10] Okubo SR, Kanawati A, Richards MW, Childress S. Evaluationof visual and instrument shade matching. J Prosthet Dent1998;80:642–8.

[11] Hassel AJ, Grossmann AC, Schmitter M, Balke Z, Buzello AM.Interexaminer reliability in clinical measurement of L*C*h*

d e n t a l m a t e r i a l s 2 5 ( 2 0 0 9 ) 276–281 281

values of anterior teeth using a spectrophotometer.Int J Prosthodont 2007;20:79–84.

[12] Wasson W, Schuman N. Color vision in dentistry.Quintessence Int 1992;23:349–53.

[13] Paul S, Peter A, Pietrobon N, Hämmerle CHF. Visual andspectrophotometric shade analysis of human teeth. J DentRes 2002;81:578–82.

[14] Jarad FD, Russell MD, Moss BW. The use of digital imagingfor colour matching and communication in restorativedentistry. Br Dent J 2005;199:43–9.

[15] Paravina RD. Evaluation of a newly developed visualshade-matching apparatus. Int J Prosthodont 2002;15:528–34.

[16] Ishikawa-Nagai S, Ishibashi K, Tsuruta O, Weber H-P.Reproducibility of tooth color gradation using a computercolor-matching technique applied to ceramic restorations. JProsthet Dent 2005;93:129–37.

[17] Dozic A, Kleverlaan CJ, El-Zohairy A, Feilzer AJ, Khashayar G.Performance of five commercially available toothcolor-measuring devices. J Prosthodont 2007;16:93–100.

[18] Ruyter IE, Nilner K, Moller B. Color stability of dentalcomposite resin materials for crown and bridge veneers.Dent Mater 1987;3:246–51.

[19] Johnston WM, Kao EC. Assessment of appearance match byvisual observation and clinical colorimetry. J Dent Res1989;68:819–22.

[20] Douglas RD, Brewer JD. Acceptability of shade differences inmetal ceramic crowns. J Prosthet Dent 1998;79:254–60.

[21] Yuan JC, Brewer JD, Monaco Jr EA, Davis EL. Defining anatural tooth color space based on a 3-dimensional shadesystem. J Prosthet Dent 2007;98:110–9.

[22] Culpepper WD. A comparative study of shade-matchingprocedures. J Prosthet Dent 1970;24:166–73.

[23] Donahue JL, Goodkind RJ, Schwabacher WB, Aeppli DP.Shade color discrimination by men and women. J ProsthetDent 1991;65:699–703.

[24] Barna GJ, Taylor JW, King GE, Pelleu GB. The influence ofselected light intensities on color perception within thecolor range of natural teeth. J Prosthet Dent 1981;46:450–3.

[25] Brewer JD, Wee A, Seghi R. Advances in color matching. DentClin N Am 2004;48:341–58.

[26] Meireles SS, Demarco FF, Santos IS, Dumith SC, Della BonaA. Validation and reliability of visual assessment with ashade guide for tooth color classification. Oper Dent2008;33:117–22.

[27] Saleski CG. Color, light and shade matching. J Prosthet Dent1972;27:263–8.

[28] ASTM D-1729-89. Standard practice for visual evaluation ofcolor differences of opaque materials. Annual book of ASTMstandards. Philadelphia, PA; 1989.