Color match of machinable lithium

disilicate ceramics: Effects of cement

color and thickness
Eva Niu, DDS, MS,a Marcus Agustin, PhD,b and
R. Duane Douglas, DMD, MSc
Southern Illinois University School of Dental Medicine, Alton, Ill; Shin
Kong Wu Ho-Su Memorial Hospital, Taipei, Taiwan
Statement of problem. The effects of cement color and thickness on the color of machinable lithium disilicate (MLD) ceramic
luted on metal foundation restorations is unknown.

Purpose. The purpose of this study was to evaluate the effects of cement color and thickness on the shade matching of MLD
restorations luted on silver-palladium (Ag-Pd) foundations.

Material and methods. Fifteen 1.5-mm-thick ceramic specimens were made from shade A1 LT lithium disilicate blocks. Five resin
cements with different colors and opacities (Multilink Automix white opaque, Multilink Automix yellow, Nexus3 white opaque,
Nexus3 white, Nexus3 yellow) of 3 thicknesses (300 mm, 100 mm, 50 mm) were sequentially luted to a roughened Ag-Pd alloy
foundation restoration block. Five shade measurements were made with a portable spectrophotometer after optically connecting the
ceramic specimen to each cement foundation block. The color differences (DE) between each specimen assembly and the target
block (a 14
14
12-mm shade A1 LT crystalized e.max block) were recorded with the CIE (Commission internationale de l’éclairage)
L*a*b* system. Clinical significance was determined by comparing color differences to perceptibility and acceptability thresholds
with the 1-sample t test (a¼.05).

Results. Both cement color and cement thickness significantly affected the mean values of the color difference (DE) of lithium
disilicate ceramic assemblies to the target block (P<.001). Among the 5 cements tested, the lowest mean DE was observed for Nexus3
white opaque cement. Overall, the combination of Nexus3 white opaque cement and 100-mm thickness led to the lowest mean DE.
The means of DE were below the clinically perceptible level (DE<2.6) for combinations of Nexus3 white opaque of 100-mm and 300-
mm cement thicknesses (P<.001 and P¼.022, respectively). The means of DE were below the clinically acceptable level (DE<5.5) for
the following combinations: Nexus3 white opaque of 50-mm thickness, Nexus3 white, and Nexus3 yellow for all tested cement
thicknesses and Multilink white opaque or Multilink yellow of either 50-mm or 100-mm cement thickness (P<.001). The means of DE
were above the clinically acceptable level (DE>5.5) for Multilink white opaque and Multilink yellow of 300-mm cement thickness.

Conclusions. The colors of machinable lithium disilicate ceramic restorations placed on Ag-Pd foundation were affected by
both the color and thickness of cements. Among the 5 cements tested in this study, Nexus3 white opaque cement of 100 mm
or 300 mm yielded shade matches below the clinical perceptible threshold (DE<2.6) relative to the target block. (J Prosthet
Dent 2014;111:42-50)

Clinical Implications
White opaque cements are more effective in masking the dark color
of silver-palladium alloy and improve the resultant shade match
of machinable lithium disilicate restorations. Increasing the cement
thickness above 100 mm did not improve the shade match. Nexus3
white opaque at all thicknesses and Multilink white opaque at 50 mm
resulted in the best shade matches.

a
Assistant Professor, Department of Restorative Dentistry, Section of Fixed Prosthodontics, Southern Illinois University, School
of Dental Medicine.
b
Professor, Department of Mathematics and Statistics, Southern Illinois University, Edwardsville.
c
Associate Professor, Chair, Department of Restorative Dentistry; and Section Head, Fixed Prosthodontics, Southern Illinois
University School of Dental Medicine.

The Journal of Prosthetic Dentistry Niu et al

January 2014
Matching the shade of ceramic res-
torations to the natural dentition is a
challenge in dentistry. Having no sup-
porting metal substructure, ceramic
restorations achieve light scattering and
transmission that are similar to that of
natural teeth.1,2 However, the trans-
lucency of ceramic restorations may be
detrimental to the perceived color when
these restorations are luted over dis-
colored dentin or metal foundation
restorations.3-14
The shade perceived for ceramic
restorations is affected by illuminants,
types of ceramic, thickness of the
ceramic materials, shades of the cores or
internal tooth structures, staining, sur-
face textures, polishing, and color and
opacities of luting agents.3-20 When a
ceramic crown is placed on an
endodontically treated tooth with a
metal post and core, the dark color and
opacity of metal foundation restorations
may lead to discoloration of the defini-
tive restorations.8-12 Among the alloys
for cast posts and cores, gold-colored
alloys cause a negligible effect on the
color of semitranslucent glass ceramic
when ceramic thickness is 1.5 mm or
more.3,9,11,12 However, because of the
increased cost of gold, alloys such as
silver-palladium (Ag-Pd) and base
metals are now commonly used for cast
posts and cores. Compared with a gold
alloy, these alloys lack a yellow hue and
are somewhat darker and bluish.
Therefore, they may substantially alter
the resultant color of semitranslucent
ceramic restorations.8,11,12,21,22
Because colors of a foundation cannot
always be chosen by clinicians, ceramic
types and thickness and cement color and
thickness become important variables in
manipulating the color of ceramic resto-
rations. The masking effect of a ceramic
material depends on its translucency,
which is determined by the material itself
and the ceramic thickness.1-5,7,9-17 High-
strength ceramic substructure materials
such as alumina- or zirconia-based ce-
ramics have high opacity and require
translucent veneering porcelain to achieve
adequate shade matches.1,2,23,24 Among
the semitranslucent glass-ceramic sys-
tems, lithium disilicate has gained
Niu et al
popularity for both anterior and posterior
crowns because of its superior esthetics,
strength, abrasion behavior, and chemical
durability.2,24-31
Previous studies have demonstrated
that increasing ceramic restoration
thickness to 2.0 mm or more is effective
in masking the influence of underlying
discolored tooth structure or metal
foundation material.3,13-17 However, in
many clinical situations, achieving 2.0
mm of axial reduction is not possible
without encroaching on the pulp and
compromising the strength of the
remaining tooth structures. In addition,
such facial reduction results in a prepa-
ration that is thin incisally. On such
preparations, overmilling is common
and can result in the compromised sta-
bility of restorations.32 When it is not
possible to attain sufficient ceramic
thickness to mask underlying metal
cores, the manipulation of the color and
thickness of cement may be the only
variable available to clinicians to mask
the effect of the underlying core on the
color of the definitive restorations.32-38
However, there is no agreement in the
literature regarding the predictability of
cement to modify the color of ceramic
restorations. Only 1 study was identified
that evaluated the interrelationship be-
tween the effect of cement thickness and
its ability to mask discolored abut-
ments.13 This study did not evaluate the
effects of metal foundations.
Spectrophotometers are among the
most accurate instruments for
measuring color in dentistry.39-41 These
instruments measure spectral reflectance
and can express it in terms of 3 coordi-
nate values (L*, a*, b*), which locate
the object’s color within the CIE (Com-
mission internationale de l’éclairage)
-L*a*b* color space. The L* coordinate
represents the brightness of an object,
the a* value represents the red (positive
value) or green (negative value) chro-
macity, and the b* value represents the
yellow (positive value) or blue (negative
value) chromacity.39-42 The color differ-
ence (DE) of 2 objects can then be
determined by comparing the diferences
between respective coordinate values of
each object.39 A lower DE value means a
43
closer color match between a specimen
and a target color. The threshold DE
value, which is perceptible to half of the
dentist observers is 2.6 DE units, and the
acceptability tolerance for half of the
dentist observers is 5.5 DE units.41
Compared with subjective visual obser-
vations, spectrophotometers offer a 33%
increase in accuracy and a more objec-
tive match in 93.3% of patients.39,43-46
The Vita EasyShade Compact (Vident)
is a small, cordless, portable intraoral
dental spectrophotometer used as an
intraoral and extraoral dental shade
matching device. Recently, it has been
used for dental shade matching
research, and several studies have sup-
ported its reliability and accuracy.43-46
The purpose of this study was to
evaluate the effects of cement color and
thickness on the resultant color of
machinable lithium disilicate (MLD)
specimens on Ag-Pd metal cores. Three
null hypotheses were evaluated. First, the
color of MLD ceramic restorations would
not be affected by the color of the cement
used. Second, the color of MLD ceramic
restorations would not be affected by the
cement thickness. Third, no interactions
would be found between the cement
color and thickness and the resultant
color of MLD restorations.
MATERIAL AND METHODS
Fifteen 1.5-mm-thick ceramic speci-
mens were made from low-translucent
shade A1 lithium disilicate computer-
aided design blocks (IPS e.max CAD LT;
Ivoclar Vivadent AG). The sample size
was determined by the authors’ unpub-
lished pilot study with try-in paste. No
power analysis was performed. The
specimens were made sectioning
ceramic blocks with a slow speed saw
(Isomet 1000 Saw; Buehler) with a 0.6-
mm-thick diamond blade. Each
sectioned slice was measured with a
digital micrometer (IP-65 Digimatic
Micrometer; Mituoyo) and adjusted to a
thickness of 1.5 mm 5 mm with 400
grit SiC paper. The sectioned slices were
polished with SiC paper (600 and 800
grit). The color-measuring surfaces of
the ceramic specimens were finely
44 Volume 111 Issue 1
Table I. Cements used
Cement
Manufacturer Colors Thickness Batch Lot No.

Multilink Ivoclar Vivadent AG White opaque (MWO); 50, 100, 300 mm MWO: N74838;
automix yellow (MLY) MLY: R06513

Nexus3 Kerr Corp White opaque (NWO); 50, 100, 300 mm NWO: 4458298;
white (NXW); yellow (NXY) NXW: 4464989; NXY: 4464987

Vivadent AG) according to the manu-

Photopolymerizing
facturer’s recommendations.
light Thimble of A foundation block was made by
digital investing a 12
14
2-mm acrylic resin
micrometer pattern (Duralay; Reliance Dental) with
Ceramic phosphate-bonded investment (Ceram-
Cement
Ag-Pd alloy block igold; Whip Mix Corp) and was cast in
Ag-Pd alloy (Albacas; Jelenko) with an
Thimble of
digital induction casting machine (Fornax T;
micrometer BEGO Dental), according to the alloy
manufacturer’s recommendations. One
Under magnification flat surface of the Ag-Pd foundation
Immersion
block was roughened with 200 grit SiC
oil
paper, rinsed with ethanol, and air-
Cement
dried. Five different color cements from
2 manufacturers were used in the
Spectrophotometer study (Table I). The cement layer was
sequentially bonded to the roughened
Ceramic surface of the Ag-Pd foundation block.
Ag-Pd alloy block Cement thicknesses of 300, 100, and 50
Custom-made specimen aligning index mm were controlled by pressing the sur-
1 Diagram of specimen fabrication and specimen face of a cover glass with the flat end of
assemblies. A, Cement thickness was controlled after photo- a digital micrometer (Digital micrometer
polymerizing cement through cover glass. B, Then, specimen IP-65; Mitutoyo) after loading the mixed
assemblies were placed on custom-made specimen aligning cements onto the foundation block. The
index for color measurements. cement was completely polymerized
with a photopolymerizing machine
Table II. DE of specimen assemblies to target block with different colors and (Optilux 501; Kerr Corp) over the cover
cement thickness glass for 90 seconds. The cover glass
Reference was then carefully separated from the
cement, which left the cement intact on
(without
the foundation block (Fig. 1).
cement) MWO MLY NWO NXW NXY The ceramic specimen was then
optically connected to the cement layer
50 mm 2.94 (0.24) 4.68 (0.18) 2.67 (0.18) 4.29 (0.18) 4.74 (0.17)
with a drop of immersion oil (immer-
100 mm 4.82 (0.18) 3.41(0.33) 4.71 (0.20) 2.21 (0.30) 3.89 (0.16) 4.63 (0.19)
sion oil type A, refractive index¼1.515;
300 mm 7.56 (0.29) 7.30 (0.21) 2.44 (0.28) 4.08 (0.15) 4.09 (0.16) Nikon) without actual bonding to the
MWO, Multilink white opaque; MLY, Multilink yellow; NOW, Nexus3 white opaque; NXW, Nexus3 ceramic surface. Shade measurements
white; NXY, Nexus3 yellow. were performed by positioning the flat
Values are given as mean (SD).
surface of a portable spectrophotom-
polished with 1200 grit SiC paper. One mandrel and served as the target block eter (Vita EasyShade Compact; Vident)
A1 LT e.max CAD block (C14, for the study. Ceramic specimens and against the center of the flat ceramic
12
14
14 mm) was polished with the target block were crystallized in a surface of each specimen assembly.
1200 grit SiC paper after removing the ceramic furnace (Programat CS; Ivoclar After 5 shade measurements with the

The Journal of Prosthetic Dentistry Niu et al

January 2014 45
Table III. DL, Da, Db of specimen assemblies to target block with different colors and cement thickness
Reference
(without cement) MWO MLY NWO NXW NXY

DL, mm
50 -1.22 (0.13) -3.54 (0.24) -0.87 (0.29) -3.08 (0.26) -3.43 (0.29)
100 -3.14 (0.24) 3.33 (0.32) -3.72 (0.25) -0.03 (0.46) -2.86 (0.22) -3.24 (0.29)
300 6.74 (0.35) -4.88 (0.26) 2.23 (0.23) -2.72 (0.23) -2.73 (0.24)

Da, mm
50 -0.50 (0.04) 0.29 (0.07) -0.04 (0.06) 0.25 (0.07) 0.29 (0.06)
100 0.12 (0.07) -0.60 (0.06) 0.18 (0.07) -0.17 (0.06) 0.40 (0.05) 0.36 (0.06)
300 -0.21 (0.06) -0.30 (0.04) -0.31 (0.07) 0.01 (0.06) 0.04 (0.05)

Db, mm
50 -2.62 (0.26) -3.01 (0.19) -2.41 (0.22) -2.96 (0.17) -3.25 (0.18)
100 -3.63 (0.28) 0.31 (0.27) -2.89 (0.16) -2.15 (0.27) -2.60 (0.16) -3.26 (0.20)
300 3.41 (0.19) -5.41 (0.17) -0.91 (0.34) -3.04 (0.18) -3.03 (0.20)

MWO, Multilink white opaque; MLY, Multilink yellow; NWO, Nexus3 white opaque; NXW, Nexus3 white; NXY, Nexus3 yellow.
Values are given as mean (SD).

Table IV. Results of 2-way ANOVA for mean DE values of combinations of
different cement color and cement thicknessa
Sum of
Source Squares
Cement used 235
Cement thickness 82.2
Interaction 186
Error 10.3
Total 514
a
S¼0.22; R ¼ 97.99%; R (adjusted)¼97.86%.
2 2
device, the next ceramic specimen was
placed on the cement surface after
adding a drop of immersion oil. After
the color of the 15 ceramic-cement-
foundation assemblies was measured,
the edge of the cement layer was loos-
ened with a chisel and cleanly peeled
off from the foundation block with
tweezers. The foundation block was
evaluated under magnification (
4.5)
to ensure a clean cement-bonding sur-
face. The next designated thickness of
cement was bonded to the foundation
block as previously described.
The same procedures were repeated
for 5 cements at 3 cement thicknesses.
As a reference, the color of the ceramic
Ag-Pd foundation assembly was me-
asured without cement by optically
connecting the ceramic specimens to the
Mean
df Square F P
4 58.8 1197 <.001
2 41.1 836 <.001
8 23.3 473 <.001
210 0.05
224
foundation block with immersion oil
only. Throughout the study, the shade
was measured by the same operator with
the same portable spectrophotometer.
The CIE-L*a*b* color space from each
color measurement was calculated and
recorded in terms of the 3 CIE coordi-
nate values (L*, a*, b*). Color differ-
ences between each ceramic specimen-
cement-foundation assembly and the
target block (DE) were calculated as the
average of 75 determinations (5 mea-
surements for each ceramic specimen-
cement-foundation assembly and 15
ceramic specimens with a ceramic
thickness of 1.5 mm).
Statistical analyses were performed
with statistical software (SPSS, v12.0.1;
SPSS Inc). Two-way ANOVA (a¼.05)
and the general model procedures were
performed to assess the effects of
cement color and thickness on the
resultant DE. Post hoc comparisons
were performed to group the effect of
the cements used and their thicknesses
on the resultant DE. Clinical significance
was determined by comparing color
differences to the perceptibility and
acceptability thresholds as determined
in a study by Douglas et al41 (DE>2.6
was considered clinically perceptible,
whereas DE>5.5 was considered clini-
cally unacceptable).
RESULTS
The mean values of DE and the dif-
ference of each CIE coordinate (DL, Da,
Db) between each ceramic specimen
assembly and the target block are listed
in Tables II and III. Results from the
2-way ANOVA demonstrated that both
the color and thickness of cement
significantly affected the color differences
(DE) of MLD ceramic-Ag-Pd assemblies
(P<.001). A significant interaction was
present between the color and thickness
of the cement (Table IV). The general
linear model was used to further inves-
tigate the data (Table V). The estimated
value of the color-cement thickness
of each cement was calculated by
subtracting the value of the intercept

Niu et al
46 Volume 111 Issue 1
Table V. General linear analysis for mean DE values of combinations of different color and thickness varied (Fig. 2). Also,
cement colors and thicknessesa the results from the grouping of the
means of DE from different cement
Sum of Mean
color-thickness combinations with the
Source df Squares Square F P Tukey method are presented in Table VII.
Except for the MWO of 300 mm, ceramic
Model 14 2525 180 2998 <.001
assemblies with Nexus3 white opaque
Error 1110 67 0.06
(NWO) cement (MWO-NWO) resulted
Corrected total 1124 2592 in the closest color match to the target
R ¼0.97; coefficient variance¼5.78; root mean standard of error¼0.25; standard error of mean¼4.24.
a 2 block. Among the 5 cements tested,
NWO yielded the lowest mean DE.
Table VI. Overall, the combination of Nexus3
Standard t white opaque cement and 100 mm
thickness resulted in the lowest mean DE
Parameter Estimate Error Value P
(Tables II, VIII; Fig. 2).
Intercept 2.94 0.03 104.05 <.001 For ceramic assemblies with MWO,
the lowest mean value of DE was
MLY 50 1.74 0.04 43.32 <.001
observed at 50-mm cement film thick-
NWO 50 -0.27 0.04 6.81 <.001
ness, followed by 100 mm, then 300 mm
NXW 50 1.34 0.04 33.51 <.001 (Table II, Fig. 2). Based on the results of
NXY 50 1.80 0.04 44.91 <.001 the 1-sample t tests, it was statistically
MWO 100 0.47 0.04 11.65 <.001 significant that the mean DE was below
MLY 100 1.76 0.04 44.00 <.001 the clinically perceptible level (DE<2.6)
NWO 100 -0.77 0.04 19.18 <.001
for the combinations of NWO with 100-
mm and 300-mm cement film thickness
NXW 100 0.95 0.04 23.79 <.001
(P<.001, P¼.022) (Table VIII). The
NXY 100 1.68 0.04 42.05 <.001 mean DE for all the other cement
MWO 300 4.62 0.04 115.30 <.001 and thickness combinations, except
MLY 300 4.35 0.04 108.64 <.001 for MWO and MLY with 300-mm
NWO 300 -0.50 0.04 12.50 <.001 cement thickness, was within clinically
NXW 300 1.14 0.04 28.45 <.001 acceptable levels (DE<5.5). The mean
DE was above the clinically acceptable
NXY 300 1.15 0.04 28.72 <.001
level (DE>5.5) for MWO and MLY of
MWO, Multilink white opaque; MLY, Multilink yellow; NWO, Nexus3 white opaque; NXW, Nexus3 300-mm cement thickness (P<.001).
white; NXY, Nexus3 yellow.
** Intercept: MWO 50 ¼ 2.94.
DISCUSSION
thickness combination minus ΔE for

Reference: ΔE for MWO X 50 μm

ΔE for each cement color/cement

5 This study evaluated the masking

effects of 2 commercial resin cements
4
of different colors and thicknesses when
MWO X 50 μm

3 MWO luting MLD ceramic to Ag-Pd founda-

2 MLY
NWO
1 NXW
0 NXY
–1 50 μm 300 μm
100 μm
–2
2 Means of DE in combinations of different cements and
film thicknesses subtracted from mean values of DEMWO50.
(Multilink white opaque [MWO] 50 mm) 4.68-2.94¼1.74 (Table VI). A graph was
from the original mean value. For drawn based on the results of the general
example, the estimated value of Multilink linear analysis to demonstrate the
yellow (MLY) 50 mm was calculated as changes of color differences as cement
The Journal of Prosthetic Dentistry
tion alloy. Based on the results of this
study, all 3 null hypotheses were rejec-
ted. The color differences between the
target block and the ceramic assemblies
were affected by both the color and
thickness of the resin cements, and an
interaction existed between the cement
color and thickness on the resultant
color of MLD restorations (Tables IV, V,
VI; Fig. 2).
Several studies reported that Ag-Pd
alloy and base metals significantly re-
duce the L* value of ceramic restora-
tions,8,11,18 although their effect on
Niu et al
January 2014 47
Table VII. Grouping information with Tukey method yellowness (b*) varied and the a* values
tended to be relatively unchanged. In this
Mean Mean Groupinga study, the control (the MLD ceramic-Ag-
Pd assemblies with no cement) was
MWO300 7.56 A
darker and bluer than the target block
MLY300 7.30 A
(Table III). White opaque cements
NXY50 4.74 B demonstrated better masking ability
MLY100 4.71 B than cements of other colors. Both NWO
MLY50 4.68 B and MWO cements effectively increased
NXY100 4.63 B the L* and b* values of the MLD
NXW50 4.29 C
ceramic-Ag-Pd assemblies and created
better shade matches in the tested
NXY300 4.09 C D
cement thicknesses (Tables II-VII, Fig. 2).
NXW300 4.08 C D
Compared with MLD ceramic-Ag-Pd as-
NXW100 3.90 D semblies with no cement (reference),
MWO100 3.41 E white or yellow cement with low opacity
MWO50 2.94 F (Nexus3 white-Nexus3 yellow [NXW-
NWO50 2.67 F G NXY]) caused minimal change in the L*
and b* values of the specimens and did
NWO300 2.44 G H
not improve the resultant shade match,
NWO 100 2.21 H
regardless of the cement thickness. Yel-
MWO, Multilink white opaque; MLY, Multilink yellow; NWO, Nexus3 white opaque; NXW, Nexus3 low cement with higher opacity (MLY)
white; NXY, Nexus3 yellow. significantly reduced the values of L* and
a
Means that do not share same uppercase letter are significantly different.
b* compared with the control and
further compromised the resultant color
Table VIII. One-sample t test for comparing mean DE values of different match of ceramic assemblies as cement
combinations to clinically perceptible level (2.6 DE units) and clinically acceptable thickness increased. The results of this
level (5.5 DE units) (n¼15)
study concur with those from a color
95% study of lithium disilicate ceramic with
SE Upper try-in paste, which recommended using
white opaque cement for cobalt-
Variable Mean SD Mean Bond T P
chromium (Co-Cr) foundation mate-
Test of m¼2.6 vs <2.6 rial.21 Among cements tested in this
NWO 100 2.21 0.30 0.08 2.34 -5.03 <.001 study, Nexus3 white opaque cement
(NWO) at all thicknesses resulted in the
NWO 300 2.44 0.28 0.07 2.57 -2.22 .022
best color match overall, followed by the
Test of m¼5.5 vs <5.5
MWO with cement thickness of 50 mm
NWO 50 2.67 0.18 0.05 2.75 -61.64 <.001 and 100 mm (Table VII, Fig. 2).
MWO 50 2.94 0.24 0.06 3.05 -41.44 <.001 Although the use of opaque cement
MWO 100 3.41 0.33 0.09 3.56 -24.57 <.001 to mask the dark shade from metal al-
NXW 100 3.90 0.16 0.04 3.97 -38.88 <.001 loys has been suggested,8,21,33 little in-
formation is available regarding the
NXW 300 4.08 0.15 0.04 4.15 -36.72 <.001
appropriate cement thickness to
NXY 300 4.09 0.16 0.04 4.17 -34.37 <.001
improve color match.13 A study by Vichi
NXW 50 4.29 0.18 0.05 4.37 -26.78 <.001 et al13 evaluated the effects of cement
NXY100 4.63 0.19 0.05 4.71 -18.00 <.001 thickness on the color of glass ceramic.
MLY 50 4.68 0.18 0.05 4.76 -17.24 <.001 The researchers reported that differ-
MLY 100 4.71 0.20 0.05 4.80 -15.55 <.001 ences in cement thickness (0.1 or 0.2
NXY 50 4.74 0.17 0.04 4.82 -16.98 <.001 mm) had little effect on the resultant
colors of leucite-reinforced ceramic
Test of m¼5.5 vs >5.5
restorations. However, only translucent
MLY 300 7.30 0.21 0.05 7.20 32.88 <.001
cements of different colors were used
MWO 300 7.56 0.29 0.08 7.43 27.38 <.001 in the study by Vichi et al,13 and dis-
DE, color differences; SD, standard deviation; SE, standard error; MWO, Multilink white opaque; colored substrates did not include
MLY, Multilink yellow; NWO, Nexus3 white opaque; NXW, Nexus3 white; NXY, Nexus3 yellow. metal foundations.
Niu et al
48
Based on the results of the current
study, white opaque cement should be
used to mask the dark color of Ag-Pd
alloy to achieve a clinically acceptable
shade match (5.5DE units). Cement
thickness for the best color match de-
pends on the color, brightness, and
opacity of the cements. Increasing the
cement thicknesses of Multilink cements
substantially affected the resultant color
of ceramic restorations, whereas, for
Nexus3, the influence of thickness was
minimal (Tables II, V, VI, VII; Fig. 2). The
mean DE values for Nexus3 white opa-
que cement were consistently lower than
for MWO cement, regardless of the
different cement thicknesses used. A
color difference below the clinical
perceptible level (DE 2.6) was achieved
for Nexus3 white opaque at 100-mm and
300-mm thickness (NWO 100 and NWO
300, respectively). For MWO cement,
the best color match was achieved at 50-
mm thickness, in which a mean color
difference for ceramic-Ag-Pd assemblies
to the target block was slightly above the
clinically perceptible level (DE¼2.94)
(Tables VII, VIII; Fig. 2). Furthermore,
increasing cement thickness from 100
mm to 300 mm presented little benefit to
the resultant color match for MLD
ceramic assemblies (Tables II, VII;
Fig. 2). In particular, for both Multilink
cements, increasing the cement thick-
ness from 100 mm to 300 mm yielded the
poorest color matches. For MWO, this
was due to its brightness and possibly
the high opacity of the cement. Although
white opaque resin cement can mask the
underlying dark color of the Ag-Pd core,
based on the results of this study,
increasing cement thickness to 300 mm
did not improve the resultant shades. In
addition, thicker cement layers have
been reported to be related to decreased
bond strength and the fracture strength
of ceramic systems.35,38
Substantial shade differences were
observed when comparing resultant
shades of ceramic-Ag-Pd assemblies
with cements of the same nominal
colors from 2 commercial manufacturers
(such as NWO vs MWO, NXY vs
MLY) (Tables II, III; Fig. 2). After the
pairwise comparison of each cement
The Journal of Prosthetic Dentistry
color-thickness combination with the
intercept group (MWO 50), the interac-
tion between cement color and thickness
was clearly demonstrated in Table IV. In
this study, for Nexus3 cements (NWO,
NXW, NXY), increasing cement thickness
from 100 to 300 mm had minimal influ-
ence on the resultant colors of MLD
ceramic-Ag-Pd assemblies. However, for
Multilink cements (MWO, MLY),
increasing cement thickness from 100 mm
to 300 mm led to an increase in that color
difference from 3.40 to 7.56 DE units
and 4.71 to 7.30 DE units (Table III).
These changes associated with an
increased cement thickness created a
resultant shade that exceeded the clini-
cally acceptable threshold of 5.5 DE
units. Therefore, with Multilink cements,
precise control of cement thickness below
100 mm is imperative for esthetic results.
In patients in whom a uniform cement
thickness under 100 mm cannot be
attained (facial undercuts from previous
restorations), the use of a cement such as
NWO may be indicated because its
resultant color is minimally affected by
varying thickness. These observations
concurred with the findings of a previous
study, which reported that color and
opacity varied among different compos-
ite luting cements of like nominal
shades.36 If Multilink cements are used to
prepare facial undercuts, then blocking
out the undercut on the tooth or foun-
dation with composite resin before digi-
tal or conventional impression taking is
imperative for uniform color results. Us-
ing try-in paste before final cementation
may be worthwhile, although 1 study
has questioned the true agreement in
color between try-in pastes and the
corresponding cements.47
This study has several limitations.
First, only a ceramic thickness of 1.5
mm was tested. Previous studies
demonstrated that the color of under-
lying substrate structures and cement
have minimal influence on the resultant
color if the ceramic thickness is 2 mm
or more, whereas the same factors
substantially alter the color of ceramic
restorations when the ceramic thickness
is 1 mm.3,7,9,13 Second, the study
design did not include actual bonding
Volume 111 Issue 1
of the luting agent to the ceramic
specimens but only to the foundation
block. To simulate a ceramic-cement-
foundation lamination, the bonded
cement-foundation complex was opti-
cally connected to the ceramic with
immersion oil. However, pilot testing of
this methodology demonstrated that
the use of oil did not affect the color of
the ceramic assemblies. Third, a
portable spectrophotometer was used
to measure colors. Although this spec-
trophotometer has been used in
numerous color studies, it is not
considered as accurate as a spectror-
adiometer.41 Color measurements of
translucent materials made with a small
window spectrophotometer or color-
imeter can be subject to deviations
caused by the edge-loss effect.39,41 To
minimize the impact of edge loss, the
ceramic specimens for the study were
made at a large dimension (12
14
mm) with flat ground surfaces. If edge
loss affected the results, these errors
would be equal for all specimens
because the same ceramic specimens
were used throughout the study and,
therefore, would have no effect on
data derived from differential color
computations.
Only shade A1 LT ceramic was used
in this study, and the results may not
be applicable to other shades and
translucencies of MLD ceramics.
Different shades of the same ceramic
with the same thickness can present
differences in light transmission,
which results in differences in optical
performance and color parameters.19
How the resultant shades of other
translucencies and shades of MLD ce-
ramics are affected by the variables
examined in this study requires further
investigation.
CONCLUSION
Within the limits of this study, the
following conclusions were drawn:
1. The color of MLD ceramic speci-
mens mock luted on Ag-Pd foundations
was affected by both the color and the
thickness of resin cements.
Niu et al
January 2014
2. With the exception of 300-mm
MWO cement, white opaque cements
yielded better shade matches than the
other color cements tested. The optimal
cement thickness for color match varied
among cements of different opacities
and colors.
3. Among the 5 cements tested in
this study, the lowest mean DE was
observed for NWO cement. The com-
bination of NWO cement at 100-mm
cement-film thickness resulted in the
best color match of ceramic assemblies
with the A1 target block.
4. Increasing the cement thickness of
white opaque cement from 100 mm to
300 mm did not improve the resultant
shade match.
REFERENCES
1. Hefferman MJ, Aquilino SA, Diaz-Arnold AM,
Haselton DR, Stamford CM, Vargas MA.
Relative translucency of six all-ceramic
systems. Part I: core materials. J Prosthet
Dent 2002;88:4-9.
2. Kelly JR, Nishimura I, Campbell SD. Ceramics
in dentistry: historical roots and current
perspectives. J Prosthet Dent 1996;75:18-32.
3. Chaiyabutr Y, Kois JC, LeBeau D,
Nunokawa G. Effects of abutment color,
cement color, and ceramic thickness on
the resulting optical color of a CAD/
CAM glass-ceramic lithium disilicate-
reinforced crown. J Prosthet Dent
2011;105:83-90.
4. Azer SS, Ayash GM, Johnston WM,
Khalil MF, Rosenstiel SF. Effect of esthetic
core shades on the final color of IPS
Empress all-ceramic crowns. J Prosthet Dent
2006;96:397-401.
5. Carossa S, Lombardo S, Pera P, Corsalini M,
Rastello ML, Preti G. Influence of posts and
cores on light transmission through different
all-ceramic crowns: spectrophotometric and
clinical evaluation. Int J Prosthodont
2001;14:9-14.
6. Goracci C, Ferrari M. Current perspectives on
post systems: a literature review. Aust Dent J
2011;56(suppl 1):77-83.
7. Denissen H, Dozic A, van Waas M, Feilzer A.
Effects of 5 manipulative variables on the
color of ceramics used for computer-
generated restorations. Quintessence Int
2007;38:401-8.
8. Trushkowsky RD. Esthetic and functional
consideration in restoring endodontically
treated teeth. Dent Clin North Am 2011;55:
403-10.
9. Volpato CA, Monteiro S, de Andrada MC,
Fedel MC, Petter CO. Optical influence of the
type of illuminant, substrate and thickness of
ceramic materials. Dent Mater 2009;25:
87-93.
Niu et al
10. Li Q, Yu H, Wang YN. Spectrophotometric
evaluation of the optical influence of core build-
up composites on all-ceramic materials. Dent
Mater 2009;25:158-65.
11. Shimada K, Nakazawa M, Kakehashi Y,
Matsumura H. Influence of abutment mate-
rials on the resultant color of heat-pressed
lithium disilicate ceramic. Dent Mater J
2006;25:20-5.
12. Nakamura T, Saito O, Fuyikawa J,
Ishigaki S. Influence of abutment substrate
and ceramic thickness on the color of heat-
presses ceramic crowns. J Oral Rehabil
2002;29:805-9.
13. Vichi A, Ferrari M, Davidson CL. Influence of
ceramic and cement thickness on the
masking of various types of opaque posts.
J Prosthet Dent 2000;83:412-7.
14. Douglas RD, Przybylska M. Predicting por-
celain thickness required for dental shade
matches. J Prosthet Dent 1999;82:143-9.
15. Azer SS, Rosenstiel SF, Seghi RR,
Johnston WM. Effect of substrate shades on
the color of ceramic laminate veneers.
J Prosthet Dent 2011;106:179-83.
16. Alghazzawi TF, Lemons J, Liu PR, Essig ME,
Janowski GM. Evaluation of the optical
properties of CAD-CAM generated yttria-
stabilized zirconia and glass-ceramic
laminate veneers. J Prosthet Dent 2012;107:
300-8.
17. Dozic A, Kleverlaan CJ, Meegdes M,
ven der Zel J, Feilzer AJ. The influence of
porcelain layer thickness on the final shade
of ceramic restorations. J Prosthet Dent
2003;90:563-70.
18. Barath VS, Faber FJ, Westland S,
Niedermeier W. Spectrophotometric analysis
of all-ceramic materials and their interaction
with luting agents and different backgrounds.
Adv Dent Res 2003;17:55-60.
19. Peixoto RT, Paulinelli VM, Sander HH,
Lanza MD, Cury LA, Poletto LT. Light
transmission through porcelain. Dent Mater
2007;23:1363-8.
20. Wang H, Xiong F, Zhenhua L. Influence
of varied surface texture of dentin
porcelain on optical properties of porcelain
specimens. J Prosthet Dent 2011;105:242-8.
21. Ge J, Wang XZ, Feng HL. Influence of
different post core materials on the color of
Empress 2 full ceramic crowns. Chin Med J
(Engl) 2006;119:1715-20.
22. Stevenson B, Ibbetson R. The effect of the
substructure on the color of samples/resto-
rations veneered with ceramic: a literature
review. J Dent 2010;38:361-8.
23. Vichi A, Louca C, Corciolani G, Ferrari M.
Color related to ceramic and zirconia
restorations: a review. Dent Mater 2011;
27:97-108.
24. Spear F, Holloway J. Which all-ceramic
system is optimal for anterior esthetics? J Am
Dent Assoc 2008;139(suppl):19-24S.
25. Shiraishi T, Wood DJ, Shinozaki N,
van Noort R. Optical properties of base
dentin ceramics for all-ceramic restorations.
Dent Mater 2011;27:165-72.
26. Ahn JS, Lee YK. Difference in the translucency
of all-ceramics by the illuminant. Dent Mater
2008;24:1539-44.
49
27. Shokry TE, Shen C, Elhosary MM,
Elkhodary AM. Effect of core and veneer
thicknesses on the color parameters of two
all-ceramic systems. J Prosthet Dent 2006;
95:124-9.
28. Brodbelt RH, O’Brien WJ, Fan PL. Trans-
lucency of dental porcelains. J Dent Res
1980;59:70-5.
29. Holand W, Schweiger MS, Frank M,
Rheinberger V. A comparison of the
microstructure and properties of IPS
Empress 2 and the IPS Empress glass-
ceramic. J Biomed Mater Res 2000;53:
297-303.
30. Otto T, De Nisco S. Computer-aided
direct ceramic restorations: 10-year pro-
spective clinical study of CEREC CAD/
CAM inlays and onlays. Int J Prosthodont
2002;15:122-8.
31. Herrguth M, Wichmann M, Reich S. The
esthetics of all-ceramic veneered and mono-
lithic CAD/CAM crowns. J Oral Rehabil
2005;32:747-52.
32. Bindl A, Windisch S, Mormann WH.
Full-ceramic CAD/CAM anterior crowns
and copings. Int Comp Dent 1999;2:
97-111.
33. Dozic A, Tsagkari M, Khashayar G,
Aboushelib M. Color management of porcelain
veneers: influence of dentin and resin cement
colors. Quintessence Int 2010;41:567-73.
34. Alqahtani MQ, Aljurais RM, Alshaafi MM.
The effects of different shades of resin luting
cement on the color of ceramic veneers. Dent
Mater J 2012;31:354-61.
35. Manso AP, Silva NR, Bonfante EA, Pegoraro TA,
Dias RA, Carvalho RM. Cements and adhesives
for all-ceramic restorations. Dent Clin North Am
2011;55:311-32.
36. Chang J, Da Silva J, Sakai M, Kristiansen J,
Nagai SI. The optical effect of composite
luting cement on all ceramic crowns. J Dent
2009;37:937-43.
37. De Azevedo Cubas GB, Camacho GB,
Demarco FF, Pereira-Cenci T. The effect of
luting agents and ceramic thickness on the
color variation of different ceramic against a
chromatic background. Eur J Dent 2011:
245-52.
38. Cekic-Negas I, Canay S, Sahin E. Bonding
of resin core materials to lithium disilicate
ceramics: the effect of resin cement film
thickness. Int J Prosthodont 2010;23:
469-71.
39. Chu SJ, Trushkowsky RD, Paravina RD.
Dental color matching instruments and sys-
tems. Review of clinical and search aspects.
J Dent 2010;38s:e2-16.
40. Öngül D, Şermet B, Balkaya MC. Visual and
instrumental evaluation of color match ability
of 2 shade guides on a ceramic system.
J Prosthet Dent 2012;108:9-14.
41. Douglas RD, Steinhauer TJ, Wee AG.
Intraoral determination of the tolerance of
dentists for perceptibility and acceptability
of shade mismatch. J Prosthet Dent
2007;97:200-8.
42. CIE (Commission Internationale de I’Eclair-
age). Colorimetry-technical report. CIE pub.
no. 15: 2004. 3rd ed. Vienna: Bureau Central
de la CIE; 2004. p. 18.
50
43. Lehmann KM, Devigus A, Igiel C,
Wentaschek S, Azar MS, Scheller H. Repeat-
ability of color-measuring devices. Eur J
Esthet Dent 2011;6:428-35.
44. Llena C, Lozano E, Amengual J, Forner L.
Reliability of two color selection
devices in matching and measuring tooth co-
lor. J Contemp Dent Pract 2011;12:19-23.
45. Khashayar G, Dozic A, Kleverlaan CJ,
Feilzer AJ. Data comparison between two
dental spectrophotometers. Oper Dent
2012;37:12-20.
46. Lasserre JF, Pop-Ciutrila IS, Colosi HA.
A comparison between a new visual
method of color matching by intraoral
camera and conventional visual and
spectrometric methods. J Dent
2011;39(suppl 3):e29-36.
47. Akgungor G, Akkayan B, Gaucher H. Influ-
ence of ceramic thickness and polymerization
mode a resin luting agent on early bond
strength and durability with a lithium dis-
ilicate ceramic system. J Prosthet Dent
2005;94:234-41.
Volume 111 Issue 1
Corresponding author:
Dr Yi-Hua (Eva) Niu
Shin Kong Wu Ho-Su Memorial Hospital
Department of Dentistry
95, Wen Chang Road, Shih Lin district, Taipei
TAIWAN
E-mail: niueva@gmail.com
Copyright ª 2014 by the Editorial Council for
The Journal of Prosthetic Dentistry.

Access to The Journal of Prosthetic Dentistry Online is reserved for print subscribers!

Full-text access to The Journal of Prosthetic Dentistry Online is available for all print subscribers. To activate your individu-
al online subscription, please visit The Journal of Prosthetic Dentistry Online. Point your browser to http://www.journals.
elsevierhealth.com/periodicals/ympr/home, follow the prompts to activate online access here, and follow the instruc-
tions. To activate your account, you will need your subscriber account number, which you can find on your mailing
label ( note: the number of digits in your subscriber account number varies from 6 to 10). See the example below in
which the subscriber account number has been circled.
Sample mailing label

This is your subscription

account number

Personal subscriptions to The Journal of Prosthetic Dentistry Online are for individual use only and may not be trans-
ferred. Use of The Journal of Prosthetic Dentistry Online is subject to agreement to the terms and conditions as indicated
online.

The Journal of Prosthetic Dentistry Niu et al