6
Coverage error of commercial skin pigments as compared to human facial skin tones Elizabeth Hungerford a , Mark W. Beatty a,b, *, David B. Marx c , Bobby Simetich a , Alvin G. Wee d a University of Nebraska Medical Center College of Dentistry, 40th & Holdrege, Lincoln, NE 68583-0740, USA b VA Nebraska-Western Iowa Healthcare System, Omaha, NE 68105, USA c Department of Statistics, University of Nebraska-Lincoln, 340 Hardin Hall, Lincoln, NE 68583-0963, USA d Creighton University School of Dentistry, 2500 California Plaza, Omaha, NE 68178, USA 1. Introduction Craniomaxillofacial defects arise from congenital malforma- tion, severe head trauma or radical surgery for treatment of head and neck cancer. Rising numbers of craniomaxillofacial defect cases are attributed to non-fatal injuries that now comprise 30% of all battlefield injuries 1,2 and increases in both the incidence and survivability of head and neck cancer. 3 Reconstructive surgery often cannot correct these defects, and j o u r n a l o f d e n t i s t r y x x x ( 2 0 1 3 ) x x x . e 1 x x x . e 6 a r t i c l e i n f o Article history: Received 9 April 2013 Received in revised form 15 July 2013 Accepted 18 July 2013 Available online xxx Keywords: Maxillofacial elastomer Prosthesis colouring Colour differences Aesthetics a b s t r a c t Objectives: It is unknown if present-day pigments used for intrinsic colouration of maxillo- facial prostheses are representative of human facial skin tones. This study’s purpose was to measure L*a*b* values of pigmented elastomers coloured by eleven skin tone pigments and determine coverage error (CE) when the pigments were compared to human facial lip and nose colour data. Methods: 11 skin tone pigments were combined at 0.1%, 1% and 10% by weight with A-2186 elastomer (n = 3). L*a*b* values were measured with a spectrophotometer and group means were used to calculate DE* colour differences with each L*a*b* value obtained for human nose and lip. Pigmented elastomer CEs were calculated for nose and lip. Results were compared to CEs for proposed shade guide colours obtained from clustering analyses of facial skin colours. Results: L* values of pigmented elastomers generally were higher than those measured for nose and lip, whereas a* values were lower. CEs for pigmented elastomers were higher than those obtained from the proposed shade guide obtained from clustered skin measurements. Conclusions: Overall, the current commercial elastomers appeared to be too white and not red enough to adequately match the skin tones of the subject population. Adjustments must be made to the existing pigmenting system in order to adequately match the skin colours of the study population. Clinical significance: The creation of a shade guide and a collection of intrinsic pigments representing the realm of human facial skin colours would greatly decrease the time a patient must sit while the clinician is obtaining an acceptable colour match for the silicone to be used for processing the final prosthesis, thereby increasing both patient satisfaction and clinician productivity. Published by Elsevier B.V. * Corresponding author. Tel.: +1 402 472 1261; fax: +1 402 472 6681. E-mail address: [email protected] (M.W. Beatty). JJOD-2109; No. of Pages 6 Please cite this article in press as: Hungerford E, et al. Coverage error of commercial skin pigments as compared to human facial skin tones. Journal of Dentistry (2013), http://dx.doi.org/10.1016/j.jdent.2013.07.010 Available online at www.sciencedirect.com journal homepage: www.intl.elsevierhealth.com/journals/jden 0300-5712/$ see front matter . Published by Elsevier B.V. http://dx.doi.org/10.1016/j.jdent.2013.07.010

Coverage error of commercial skin pigments as compared to human facial skin tones

  • Upload
    alvin-g

  • View
    213

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Coverage error of commercial skin pigments as compared to human facial skin tones

JJOD-2109; No. of Pages 6

Coverage error of commercial skin pigments ascompared to human facial skin tones

Elizabeth Hungerford a, Mark W. Beatty a,b,*, David B. Marx c,Bobby Simetich a, Alvin G. Wee d

aUniversity of Nebraska Medical Center College of Dentistry, 40th & Holdrege, Lincoln, NE 68583-0740, USAbVA Nebraska-Western Iowa Healthcare System, Omaha, NE 68105, USAcDepartment of Statistics, University of Nebraska-Lincoln, 340 Hardin Hall, Lincoln, NE 68583-0963, USAdCreighton University School of Dentistry, 2500 California Plaza, Omaha, NE 68178, USA

j o u r n a l o f d e n t i s t r y x x x ( 2 0 1 3 ) x x x . e 1 – x x x . e 6

a r t i c l e i n f o

Article history:

Received 9 April 2013

Received in revised form

15 July 2013

Accepted 18 July 2013

Available online xxx

Keywords:

Maxillofacial elastomer

Prosthesis colouring

Colour differences

Aesthetics

a b s t r a c t

Objectives: It is unknown if present-day pigments used for intrinsic colouration of maxillo-

facial prostheses are representative of human facial skin tones. This study’s purpose was to

measure L*a*b* values of pigmented elastomers coloured by eleven skin tone pigments and

determine coverage error (CE) when the pigments were compared to human facial lip and

nose colour data.

Methods: 11 skin tone pigments were combined at 0.1%, 1% and 10% by weight with A-2186

elastomer (n = 3). L*a*b* values were measured with a spectrophotometer and group means

were used to calculate DE* colour differences with each L*a*b* value obtained for human nose

and lip. Pigmented elastomer CEs were calculated for nose and lip. Results were compared to

CEs for proposed shade guide colours obtained from clustering analyses of facial skin

colours.

Results: L* values of pigmented elastomers generally were higher than those measured for

nose and lip, whereas a* values were lower. CEs for pigmented elastomers were higher than

those obtained from the proposed shade guide obtained from clustered skin measurements.

Conclusions: Overall, the current commercial elastomers appeared to be too white and not

red enough to adequately match the skin tones of the subject population. Adjustments must

be made to the existing pigmenting system in order to adequately match the skin colours of

the study population.

Clinical significance: The creation of a shade guide and a collection of intrinsic pigments

representing the realm of human facial skin colours would greatly decrease the time a

patient must sit while the clinician is obtaining an acceptable colour match for the silicone

to be used for processing the final prosthesis, thereby increasing both patient satisfaction

and clinician productivity.

Published by Elsevier B.V.

Available online at www.sciencedirect.com

journal homepage: www.intl.elsevierhealth.com/journals/jden

1. Introduction

Craniomaxillofacial defects arise from congenital malforma-

tion, severe head trauma or radical surgery for treatment of

* Corresponding author. Tel.: +1 402 472 1261; fax: +1 402 472 6681.E-mail address: [email protected] (M.W. Beatty).

Please cite this article in press as: Hungerford E, et al. Coverage error of

Journal of Dentistry (2013), http://dx.doi.org/10.1016/j.jdent.2013.07.010

0300-5712/$ – see front matter. Published by Elsevier B.V.http://dx.doi.org/10.1016/j.jdent.2013.07.010

head and neck cancer. Rising numbers of craniomaxillofacial

defect cases are attributed to non-fatal injuries that now

comprise 30% of all battlefield injuries1,2 and increases in both

the incidence and survivability of head and neck cancer.3

Reconstructive surgery often cannot correct these defects, and

commercial skin pigments as compared to human facial skin tones.

Page 2: Coverage error of commercial skin pigments as compared to human facial skin tones

j o u r n a l o f d e n t i s t r y x x x ( 2 0 1 3 ) x x x . e 1 – x x x . e 6e2

JJOD-2109; No. of Pages 6

a disfigured appearance produces disability through profound

quality of life issues that include a loss of self-esteem,

depression and withdrawal from society.4,5 Consequently, a

facial prosthesis that replaces missing structure and

aesthetically blends with surrounding facial features is critical

to restoring self-esteem and eventual re-integration into

everyday life.6

One of the challenges in constructing a facial prosthesis is

the determination of a shade that serves as a base colourant

within an elastomer. A common technique is to mix a

combination of pigments into a siloxane polymer and hold

the pigmented elastomer next to the face to assess the

accuracy of the mixture. This trial-and-error method is time

consuming, as it may require multiple trials to achieve the

proper shade. Needed in this process is a group of base

colourants representative of the range of facial shades present

within the human population. Pre-mixed skin shades are

marketed for this purpose, but their ranges of shades have not

previously been compared to facial skin measured in a human

population. Recently it was reported that through clustering

analyses, five base shades could be identified to represent lip

and nose skin colours for a group of 119 human subjects, and

therefore serve as a basis for a potential shade guide.7 With

these data available, the purpose of this study was to compare

human facial skin colours in this population with eleven

cosmetic pigments currently available from Factor II, Inc., as

intrinsic colourants for maxillofacial prostheses.8 CIELab

colour differences between pigmented elastomers and facial

skin measurements were used to compare coverage errors of

pigmented elastomers with those calculated for the five

shades obtained from clustered skin colours.

2. Materials and methods

This study tested a platinum-catalyzed, vinyl-terminated

poly(dimethyl siloxane) elastomer (A-2186, Factor II, Lakeside,

AZ) combined with a functional intrinsic cosmetic pigment.8

The vinyl elastomer was combined with a polymethyl

hydrogen siloxane cross-linking agent at a 10:1 ratio by

weight. Once elastomer components were combined and

thoroughly mixed, one of eleven pigments (Table 1) was added

in three different concentrations by weight: 0.1%, 1.0%, and

10%. For each concentration the sample size was three (n = 3),

Table 1 – Functional intrinsic skin pigments evaluated inthis study.

Name Product code Lot number

1. Bisque FI-SK05 S 070111

2. Blush FI-SK13 B 021511

3. Cream FI-SK07 SB 081611

4. Dark Brown FI-SK21 BS 081711

5. Dusk FI-SK17 TK 080509

6. Honey FI-SK09 BS 082510

7. Ivory FI-SK03 BT 042210

8. Mocha FI-SK19 T 042909

9. Naturelle FI-SK01 B 062311

10. Santa Fe FI-SK11 SB 052711

11. Soft Brown FI-SK15 BS 052011

Please cite this article in press as: Hungerford E, et al. Coverage error of

Journal of Dentistry (2013), http://dx.doi.org/10.1016/j.jdent.2013.07.010

thus yielding a total of 99 samples. The elastomer – pigment

combination was placed under 5 � 10�3 torr vacuum for

10 min to remove air from the system, then poured into three

disc-shaped moulds with dimension 6 mm thick-

ness � 34 mm diameter. The moulds were placed into a

convection oven and held at 74 degrees Celsius for 1 h to

achieve full polymerization. The moulds were removed and

allowed to cool to room temperature, then a specimen number

was scribed on the side of the disc with an indelible marker.

A colour reflectance spectrophotometer with a 1 cm

aperture (CM-2002, Konica Minolta) was used to measure

the colour of each elastomer disc on standard black and white

background tiles using the CIEL*a*b* colour space. The

instrument’s repeatability was DE* < 0.01 and validity com-

pared to British Ceramic Research Association colour tiles was

0.025 < DE* < 0.219. A D65 illuminant and a viewing angle of

108 were chosen. After calibration, colour was measured first

on white (L*: 97.256, a*: 0.307, b*: 2.360) and second on black (L*:

4.53, a*: �0.86, b*: �1.17) background tiles in a darkened room.

L*a*b* coordinates and spectral curve data were recorded.

Since comparisons only between pigments and human

skin colours were desired, visible differences caused by

elastomer translucency were considered unacceptable. There-

fore, during colour measurement of a test sample, the

influence of background colour for a translucent sample

had to be removed. Translucency parameter (TP) was

calculated for each disc using the equation9,10

TP ¼ ½ðLb � LwÞ2 þ ðab � awÞ2 þ ðbb � bwÞ2�1=2;

where the subscripts b and w refer to colour coordinates mea-

sured on black and white backgrounds, respectively. Group

means and standard errors of the mean (s.e.m.) were calculated

for each elastomer formulation. Since the threshold for 50:50

colour difference perceptibility has been reported to be for

approximately DE* = 1.1,11 groups demonstrating TP greater

than 1.1 were excluded from skin comparison measurements.

Skin colour measurements used for this study were taken

from those reported previously.7 Briefly, colour measurements

were recorded from the vermilion border of the lip and the tip

of the nose for 119 human subjects (IRB #2003H0001, The Ohio

State University). The population was stratified by age (18–29,

30–39, 40–49, 50–59, 60–85), gender (female, male), and race/

ethnicity (Asian/Pacific Islander, Black, White, Other). Written

informed consent was secured and skin colour measurements

were obtained with a spectroradiometer (PR 705; Photo

Research, Inc., Chatsworth, CA) and Xenon arc lamp (300 W;

Oriel Instruments, Stratford, CO) that were set up in an optical

configuration of 08 observation and 458 illumination. Reflec-

tance measurements were obtained between 380 and 780 nm

in 2 nm bandwidth intervals, and the data converted to L*a*b*

(CIELAB) colour coordinates.

For comparison of elastomers with skin colours, mean

� s.e.m. L*, a* and b* values were determined for each

elastomer group. The colour difference (DE*) between each

elastomer group mean and each skin colour measured for

nose or lip was calculated from the following equation:

DE� ¼ ½ðDL�Þ2 þ ðDa�Þ2 þ ðDb�Þ2�1=2:

For evaluation of colour differences between elastomers

and skin, minimum, maximum and average DE* values were

commercial skin pigments as compared to human facial skin tones.

Page 3: Coverage error of commercial skin pigments as compared to human facial skin tones

Table 2 – Translucency parameters (mean W s.e.m.) of 11pigments at three concentrations.

0.10% 1% 10%

Bisque 10.526 � 0.1265 0.09 � 0.0099 0.06 � 0.0037

Blush 17.485 � 0.3366 0.405 � 0.0278 0.085 � 0.0168

Cream 13.879 � 0.2152 0.213 � 0.0166 0.088 � 0.0085

Dark Brown 9.539 � 0.0979 0.07 � 0.0023 0.163 � 0.0072

Dusk 12.259 � 0.2561 0.048 � 0.0006 0.055 � 0.0202

Honey 11.564 � 0.2311 0.05 � 0.0032 0.038 � 0.0157

Ivory 10.110 � 0.1943 0.082 � 0.0250 0.084 � 0.0184

Mocha 7.466 � 0.1993 0.061 � 0.0074 0.083 � 0.0141

Naturelle 10.687 � 0.0294 0.23 � 0.0203 0.061 � 0.0066

Santa Fe 20.032 � 0.4108 0.944 � 0.0819 0.042 � 0.0061

Soft Brown 11.185 � 0.2359 0.063 � 0.0064 0.102 � 0.0102

j o u r n a l o f d e n t i s t r y x x x ( 2 0 1 3 ) x x x . e 1 – x x x . e 6 e3

JJOD-2109; No. of Pages 6

compared to those reported for the threshold of 50:50

perceptibility (DE* = 1.1) and 50:50 acceptability (DE* = 3.0).11

Coverage error was used as an index for estimating the

relative error between the pigmented elastomers and skin

colours of the human population. For each pigmented

elastomer group, mean L*a*b* values were calculated for the

three discs. DE* was calculated for colour differences occurring

between each pigmented elastomer group’s mean L*a*b* value

and the L*a*b* value for each of the 119 lip and nose

measurements. From this, the smallest DE* was determined

for each elastomer group. The average minimum DE* for the

pigmented elastomers was then calculated for lip and nose.

The coverage error (CE), was calculated by the expression12,13

CE ¼P

DEmin

n¼P

MinffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiðDL�Þ2 þ ðDa�Þ2 þ ðDb�Þ2

q

n;

where n = 119, the number of lip or nose measurements. The

coverage error, therefore, was the average difference between

a lip or nose colour and the nearest pigmented elastomer

colour. Likewise, CE was calculated for the five skin shade

cluster centroids, with CE representing the average difference

between a lip or nose colour and the nearest clustered skin

shade.

3. Results

Mean translucency parameter values of the 11 pigments at 3

concentrations are reported in Table 2. All 0.1% pigment

concentration samples exhibited translucency parameters

exceeding 3.0, which demonstrated that translucency visibly

affected colour for this concentration. Therefore, the 0.1%

specimens could not be considered for comparison with

Table 3 – Colour differences and coverage error between lip d

Pigment Concentration Mean DE*

Colour difference (DE*)

Bisque 1% 27.712

Bisque 10% 27.152

Blush 1% 26.724

Blush 10% 26.075

Cream 1% 27.183

Cream 10% 26.648

Dusk 1% 14.743

Dusk 10% 15.321

Dark Brown 1% 19.701

Dark Brown 10% 18.907

Honey 1% 24.266

Honey 10% 23.523

Ivory 1% 26.743

Ivory 10% 26.924

Mocha 1% 17.473

Mocha 10% 16.828

Naturelle 1% 30.390

Naturelle 10% 30.799

Soft Brown 1% 24.021

Soft Brown 10% 23.841

Santa Fe 1% 21.674

Santa Fe 10% 20.096

Coverage error (DE*) 13.272

Please cite this article in press as: Hungerford E, et al. Coverage error of

Journal of Dentistry (2013), http://dx.doi.org/10.1016/j.jdent.2013.07.010

human skin measurements and were removed from further

analysis. The translucency parameters for the 1.0% and 10%

specimens ranged from 0.006 to 0.9 which remained below the

50:50 threshold for perceptibility (DE* = 1.1) and could thus be

investigated further.

DE* colour differences between the 1% and 10% pigmented

elastomer groups and the lip and nose data were analyzed and

the results are presented in Tables 3 and 4. In both tables, the

range of DE* colour differences are presented as the minimum,

maximum and average DE* values for each lip and nose

comparison with the elastomers. For lip comparisons (Table

3), minimum colour differences ranged 4.793 � DE* � 15.988.

The closest colour match with a single human lip measure-

ment occurred with 10% Mocha (DE* = 4.793). Maximum DE*

values ranged 31.313 � DE* � 56.755 and coverage error ranged

3.459 � DE* � 28.094.

In Table 4, DE* comparisons between pigmented elastomers

and nose measurements demonstrated minimum colour

ata and pigmented elastomer specimens.

s.e.m. Minimum Maximum

6.737 13.756 52.578

7.035 11.169 53.418

6.808 12.290 52.123

7.299 7.976 53.701

5.853 15.553 48.871

6.135 12.959 50.016

4.019 6.668 34.094

3.842 7.540 33.556

5.205 7.832 39.505

5.241 7.120 39.516

6.244 10.612 48.697

6.307 8.599 48.826

6.233 12.179 50.636

6.408 10.765 51.753

4.825 5.516 31.968

4.817 4.793 31.313

6.734 15.988 54.759

7.104 13.955 56.755

6.280 13.115 47.606

7.237 9.446 51.157

5.809 10.951 44.378

5.824 10.064 43.895

0.348 3.459 28.094

commercial skin pigments as compared to human facial skin tones.

Page 4: Coverage error of commercial skin pigments as compared to human facial skin tones

Table 4 – Colour differences and coverage error between nose data and pigmented elastomer specimens.

Pigment Concentration Mean DE* s.e.m. Minimum Maximum

Colour difference (DE*)

Bisque 1% 21.253 7.496 8.057 42.035

Bisque 10% 20.215 7.964 6.231 41.912

Blush 1% 20.059 7.643 6.607 41.158

Blush 10% 18.721 8.351 4.737 41.148

Cream 1% 19.714 6.515 9.475 38.331

Cream 10% 18.598 7.120 7.311 38.472

Dusk 1% 9.766 4.114 1.008 19.154

Dusk 10% 10.700 4.386 0.236 19.819

Dark Brown 1% 15.216 4.618 7.434 30.180

Dark Brown 10% 14.005 4.802 6.043 29.514

Honey 1% 16.209 7.338 4.852 36.691

Honey 10% 14.850 7.645 3.101 35.905

Ivory 1% 18.461 7.327 6.886 38.722

Ivory 10% 18.262 7.667 6.361 39.138

Mocha 1% 15.925 2.924 9.959 24.087

Mocha 10% 15.275 2.959 9.190 22.906

Naturelle 1% 23.883 7.477 10.657 44.527

Naturelle 10% 23.819 7.993 9.913 45.482

Soft Brown 1% 19.678 6.444 9.090 38.565

Soft Brown 10% 18.611 7.737 5.753 40.156

Santa Fe 1% 15.369 6.182 5.972 33.821

Santa Fe 10% 12.934 6.481 4.221 32.168

Coverage error (DE*) 7.523 0.303 0.236 17.927

j o u r n a l o f d e n t i s t r y x x x ( 2 0 1 3 ) x x x . e 1 – x x x . e 6e4

JJOD-2109; No. of Pages 6

differences ranging 0.236 � DE* � 10.657, maximum differ-

ences ranging 19.154 � DE* � 45.482 and coverage errors

ranging 0.236 � DE* � 17.927. Pigmented elastomers with

colours closest to those measured for human noses were

10% dusk, 1% dusk and 10% honey (DE* = 0.236, 1.008 and 3.101,

respectively).

Based on results from clustering analyses, previous

research identified five colours capable of serving as a shade

guide for this same human population.7 Each colour repre-

sented the centroid of a population cluster. Table 5 presents

colour difference and coverage error results for the five shades

and each of the 119 lip and nose colour measurements. For

Table 5 – Colour differences and coverage error betweencluster centroids and lip and nose data.7

Cluster Mean DE* s.e.m. Minimum Maximum

Lip colour difference (DE*)

1 17.097 0.568 7.064 44.366

2 13.384 0.406 4.240 36.165

3 9.206 0.510 1.872 34.270

4 14.321 0.569 1.709 33.736

5 11.190 0.616 1.304 40.744

Lip coverage

error (DE*)

5.603 0.278 1.304 22.199

Nose colour difference (DE*)

1 9.239 0.663 1.520 30.344

2 8.557 0.402 1.121 21.367

3 14.334 0.377 2.478 22.618

4 17.727 0.556 5.976 28.805

5 12.184 0.480 3.901 28.570

Nose coverage

error (DE*)

4.759 0.174 1.121 10.769

Please cite this article in press as: Hungerford E, et al. Coverage error of

Journal of Dentistry (2013), http://dx.doi.org/10.1016/j.jdent.2013.07.010

both lip and nose measurements, three cluster centroids

demonstrated minimum DE* values less than 2.5. Coverage

error values ranged 1.304 � DE* � 22.199 for lip and

1.121 � DE* � 10.769 for nose.

To gain a better understanding of the spatial relationship of

colour coordinates among 22 pigmented elastomers, 119 lips

and noses, and 5 cluster centroids, colour data were plotted.

Results are presented in Fig. 1. Fig. 1(a) presents an overall

three-dimensional view of data, and Fig. 1(b)–(d) present two-

dimensional views along primary colour axes. Generally,

pigmented elastomers were clustered at higher L* values and

lower a* values compared to lip and nose measurements.

Cluster centroids, as expected, demonstrated L*a*b* values

that were representative of the subject population.

4. Discussion

Until now, there has been no research comparing the

commercially available intrinsic pigments with human skin

data. Previous research7 identified five colours that represent a

subject population based on clustering analysis and proposed

the use of these colours to form a shade guide for silicone

facial prostheses. This research provides colour comparisons

of intrinsic pigments both with human skin and the five

proposed shade guide colours.

As has been recognized during clinical construction of a

prosthesis, pigment concentration can affect the final appear-

ance of the elastomer. During construction, pigments are

mixed in variable amounts in order to achieve a custom match

with the surrounding facial tissues for each patient. Since the

effect of pigment concentration was unknown for this

pigmenting system, a 10-fold difference in concentration

was chosen as a starting point. Also, since the aim of this

commercial skin pigments as compared to human facial skin tones.

Page 5: Coverage error of commercial skin pigments as compared to human facial skin tones

Fig. 1 – Scatter plots of colour coordinates for 22 pigmented elastomer groups, 119 human noses, 119 human lips and 5

cluster centroids proposed as shade guide colours for this population: (a) 3D L*a*b* plot, (b) 2D a* vs. b* plot, (c) 2D L* vs. a*

plot and (d) 2D L* vs. b* plot. Due to the number of data points shown, certain symbols are not visible.

j o u r n a l o f d e n t i s t r y x x x ( 2 0 1 3 ) x x x . e 1 – x x x . e 6 e5

JJOD-2109; No. of Pages 6

research was to evaluate the pigmenting system based solely

on colour, any visually detectable contribution by translucen-

cy had to be eliminated. By measuring the translucency

parameter at each pigment concentration for a test sample of

constant thickness, it was possible to determine that

translucency contributed to the visual appearance of 0.1%

pigmented elastomers. Consequently, these samples had to be

removed from further analysis for colour comparisons with

human skin and proposed shade guide colours.

For 1.0% and 10% pigment concentrations, the results

demonstrated that the ranges of colour difference between

pigmented elastomers and human lips were higher than those

observed for human noses. Minimum colour differences

(which represented the closest matches between elastomers

Please cite this article in press as: Hungerford E, et al. Coverage error of

Journal of Dentistry (2013), http://dx.doi.org/10.1016/j.jdent.2013.07.010

and human skin) were 4.557–5.331 DE* units lower for noses,

and maximum colour differences were 11.273–12.159 DE* units

lower. Coverage errors, the average differences between lip or

nose colours and the nearest pigmented elastomer colours,

were 4.977–6.916 DE* units lower for noses than for lips.

Collectively these results demonstrated that the pigmenting

system more closely approximated human nose colours than

the human lip colours. This implies that additional pigments

are likely needed to match lip colours more closely.

Comparison of mean coverage errors for clustered cen-

troids (which represented proposed shade guide colours) with

those for pigmented elastomers demonstrated that the

centroid values were 2.764 and 8.513 DE* units lower for nose

and lip colours, respectively. Also, results displayed in Fig. 1

commercial skin pigments as compared to human facial skin tones.

Page 6: Coverage error of commercial skin pigments as compared to human facial skin tones

j o u r n a l o f d e n t i s t r y x x x ( 2 0 1 3 ) x x x . e 1 – x x x . e 6e6

JJOD-2109; No. of Pages 6

illustrate that cluster centroid positions were located closer to

human nose and lip colours than were the group of pigmented

elastomers studied here, which was expected since the

centroids were derived from observations taken from the

study population. However, Fig. 1(b) and (c) also shows that the

group of 1% and 10% pigmented elastomers represented

colours that were too low in redness (a*) or blackness (L*) to

effectively cover the study population. These results strongly

suggest that future development of more accurate intrinsic

colourants rely upon shades that represent the spectrum of

skin colours present within the human population. Future

research should focus on conducting an expanded, properly

powered clinical study capable of identifying pigment colours

that adequately span the realm of human skin colours that can

be incorporated into a usable shade guide for facial prosthesis

construction.

Although this research provided first-time information

regarding the ability of commercial pigments to cover lip and

nose colours in a study population, a number of limitations

should be noted. An obvious limitation is that the reported

results pertained only to this human population, and a larger

population is warranted for future research. A second

limitation is that the results were based on measurements

taken from only two facial locations. Future research also

should include measurements taken from other locations,

particularly forehead, cheek and ear. It is noteworthy that

different equipment was used to provide illumination and

measure colour for human skin (non-contact spectroradi-

ometer) and pigmented elastomers (spectrophotometer). This

drawback enhanced the importance of removing translucency

as a component of perceived colour. Similar future studies

should employ identical equipment for illumination and

colour measurement for skin and elastomers. It is important

to note that only one commercial pigmenting system was

evaluated here; different colour coverage parameters for

human skin may be identified for different commercial

pigmenting systems. Finally, colour measurements for pig-

mented elastomers were conducted only on freshly mixed

materials. Colour changes produced by ageing and weathering

would be expected to produce colour shifts accompanied by a

different range of coverage for human skin colour. This

phenomenon should be considered when engineering

changes into future facial prosthetic materials.

5. Conclusions

Within the limitations of this study, the following conclusions

could be drawn:

1. Elastomers containing 0.1 wt.% pigment produced trans-

parency parameters above the threshold limits for 50:50

perceptibility and acceptability, whereas elastomers

containing 1.0 and 10.0 wt.% pigment produced transpar-

ency parameters below the threshold limit for 50:50

perceptibility.

2. The commercial pigmenting system examined in this

project produced colours that were low in redness and

Please cite this article in press as: Hungerford E, et al. Coverage error of

Journal of Dentistry (2013), http://dx.doi.org/10.1016/j.jdent.2013.07.010

blackness, as compared to colours measured for lips and

noses within a human subject population.

3. Colours derived from clusters of measured skin colours

produced lower coverage errors than did colours produced

by a commercial pigmenting system.

4. A large-scale human subject study that measures colour at

several facial locations is needed for further development of

a shade guide for facial prosthetic materials.

Conflict of interest statement

The authors deny any conflict of interest and hold no

association with any of the products or manufacturers,

financial or otherwise, as stated in the manuscript.

r e f e r e n c e s

1. Owens BD, Kragh JF, Wenke JC, Macaitis J, Wade CE, HolcombJB. Combat wounds in operation Iraqi freedom and operationenduring freedom. Journal of Trauma 2008;64:295–9.

2. Lew TA, Walker JA, Wenke JC, Blackboune LH, Hale RG.Characterization of craniomaxillofacial battle injuriessustained by United States service members in the currentconflicts of Iraq and Afghanistan. Journal of Oral andMaxillofacial Surgery 2010;68:3–7.

3. Horner MJ, Ries LAG, Krapcho M, Neyman N, Aminou R,Howlader N, et al. editors. SEER cancer statistics review, 1975–2006. Bethesda, MD: National Cancer Institute; 2012. http://seer.cancer.gov/csr/1975_2006/, based on November 2008SEER data submission, posted to the SEER web site, accessedNovember 20.

4. Macgreggor FC. Some psycho-social problems associatedwith facial deformities. American Sociological Review1951;16:629–38.

5. Bailey LW, Edwards D. Psychological considerations inmaxillofacial prosthetics. Journal of Prosthetic Dentistry1975;34:533–8.

6. Roefs AJ, van Oort RP, Schaub RM. Factors related to theacceptance of facial prostheses. Journal of Prosthetic Dentistry1984;52:849–52.

7. Wee AG, Beatty MW, Gozalo-Diaz D, Kim-Pusateri S, MarxDB. Proposed shade guide for human facial skin andlip – a pilot study. Journal of Prosthetic Dentistry2013;110:82–9.

8. In: http://www.factor2.com/product_p/fi-sk.htm. [accessed 20.02.13].9. Johnston WM, Ma T, Kienle BH. Translucency parameter of

colorants for maxillofacial prostheses. International Journal ofProsthodontics 1995;8:79–86.

10. Hu X, Johnston WM. Translucency estimation for thickpigmented maxillofacial elastomer. Journal of Dentistry2011;39:e2–8.

11. Paravina RD, Majkic G, Del Mar Perez M, Kiat-amnuay S.Color difference thresholds of maxillofacial skinreplications. Journal of Prosthodontics 2009;18:618–25.

12. O’Brien WJ, Boenke KM, Groh CL. Coverage errors of twoshade guides. International Journal of Prosthodontics 1991;4:45–50.

13. Hannigan A, Lynch CD. Statistical methodology in oral anddental research: pitfalls and recommendations. Journal ofDentistry 2013;41:385–92.

commercial skin pigments as compared to human facial skin tones.