Você está na página 1de 7

The effect of tray selection, viscosity of impression material, and sequence

of pour on the accuracy of dies made from dual-arch impressions


Jeffrey A. Ceyhan, DDS, MSD,a Glen H. Johnson, DDS, MS,b and Xavier Lepe, DDS, MSc
School of Dentistry, University of Washington, Seattle, Wash
Statement of problem. Dual-arch trays are often used to generate impressions of prepared teeth and of the
opposing arch simultaneously. There is concern that accuracy of the casts generated with this technique can be
affected by the type of tray, viscosity of the impression material, and sequence of pouring the cast.
Purpose. This study compared the accuracy of working dies made from impressions with metal and plastic
dual-arch trays, for 2 different viscosities of impression tray material and by altering which side of the impression
was poured first.
Material and methods. Impressions were made of a typodont mandibular arch containing a circular stainless
steel crown preparation (standard). There were 3 variables: type of dual-arch tray, impression material viscosity,
and order of pour of the impressioned arches. A balanced design with independent samples was used (n10).
Two types of dual-arch trays, plastic (Triple Tray) and metal (COE Impression Tray), and 2 viscosities of addition
silicone for the tray were used (Aquasil Rigid and Aquasil Monophase). Type IV gypsum (Fuji-Rock) with a ratio
of 20 mL of distilled water to 100 g of powder was hand-mixed for 10 seconds then mixed under vacuum for 40
seconds and poured into the trays while being vibrated. One side of the dual-arch impression was poured with
35 g of stone and allowed to set for 1 hour before the other side was poured with 35 g of stone. The order of pour
was randomized, and all casts were allowed to set for 24 hours at room temperature before removal. The dies were
measured in 3 dimensions (buccolingual, mesiodistal, and occlusogingival) with a measuring microscope. The
gypsum working dies were placed into a custom jig fabricated to permit measurement at a fixed, reproducible
position under the microscope. Each dimension of the working dies was measured 3 times, and the mean was used
for the sample value. The same 3 aspects of the stainless steel standard were measured multiple times, before and
then at the conclusion of measuring all working dies, to arrive at the 3 standard values to which all working die
means were compared. The means for the standard used in the statistical analysis were those taken at the
conclusion of the study. The intraexaminer variation for measuring the standard was 0.001 mm. A 3-factor
analysis of variance was used for the statistical analysis with hypothesis testing at .05.
Results. Statistically significant differences were found with viscosity selection for the buccolingual and occlusogingival dimensions of the working die. The rigid material produced working dies slightly taller (1 m) than the standard,
and those from the monophase material were 4 m shorter. Regarding tray selection, metal trays were slightly more
accurate in the mesiodistal dimension, and when monophase was used in a plastic tray, gypsum dies were nearly 30 m
smaller in the mesiodistal dimension (P.05). Differences were not detected for sequence of pouring impressions.
Conclusion. Within the limitations of this study, the monophase material, when compared with the rigid
impression material, was most accurate for the occlusogingival and mesiodistal dimensions, although not as
accurate in the buccolingual. This buccolingual difference (0.002 mm-0.006 mm) would be clinically inconsequential with the application of die spacer. The rigid impression material was also unaffected by tray selection for
the mesiodistal, whereas monophase was affected. When a monophase impression material was used, plastic
dual-arch trays yielded gypsum dies which were significantly smaller (0.029 mm) than the ones generated from
the metal trays (0.006 mm). Thus rigid impression materials can be recommended for use in dual-arch trays;
however, the magnitude of the differences would generally not be clinically significant because they could be
compensated for with several coats of die spacer. (J Prosthet Dent 2003;90:143-9.)

CLINICAL IMPLICATIONS
The thickness of 1 coat of die spacer can vary from 8 to 40 m. Therefore, the differences in
accuracy between trays, viscosities, and pour techniques evaluated in this study would be clinically inconsequential. Although a rigid consistency of impression material in a dual-arch tray
yielded working dies slightly more accurate than the monophase material, clinically acceptable
impressions can be made when plastic or metal dual-arch trays are used with either a monophase
or rigid impression material.
a

Resident, Graduate Prosthodontics; current private practice, Calgary, Alberta.


b
Professor, Department of Restorative Dentistry.
c
Associate Professor, Department of Restorative Dentistry.

AUGUST 2003

THE JOURNAL OF PROSTHETIC DENTISTRY 143

THE JOURNAL OF PROSTHETIC DENTISTRY

CEYHAN, JOHNSON, AND LEPE

he dual-arch or double arch impression technique


as described by Wilson and Werrin1 is convenient in that
it makes the required maxillary and mandibular impressions, as well as the interocclusal record in 1 procedure.
The authors state that many types of impression materials can be used with the double arch trays, but a stiff
bodied impression material is preferred to ensure tray
rigidity, and more rigid trays reduce the possibility of
distortion in the impression. They also advocate that the
counterimpression (opposing side) should always be
poured first, followed by the working side (preparation
side) impression; however, there are no references that
support these claims. Several authors describe this impression technique.2-5 The accuracy of casts generated
from this technique remains in question because there is
little information available in the literature, and the protocol described by Wilson and Werrin1 is not always
followed.
Davis and Schwartz6-8 published 3 studies examining
the accuracy of dual-arch impressions. The first study
examined the accuracy of dual-arch and custom tray
impressions. Using reference marks made on a typodont, measurement of the mesiodistal, buccolingual,
and occlusogingival dimensions of the casts were made.
The working side of the dual-arch impressions was
poured first, and a regular viscosity impression material
was used in the trays. Their results show that custom
trays are less accurate than both types of dual-arch trays
in the buccolingual dimension and that the metal dualarch trays are more accurate than the other trays for the
inter-tooth distances.
The second study investigated the accuracy of second
pour casts with the 2 types of dual-arch impressions and
a full-arch custom tray impression. A double-mix impression technique was used with regular body impression material in the dual-arch trays, with the working
side of the impressions being poured first, followed by
the opposing arch 1 hour later. A statistical difference in
accuracy between the 3 tray types could not be shown
except, once again, the metal dual-arch trays were more
accurate than the other trays in the intertooth measurements.
The third study examined the marginal adaptation of
gold castings made from dual-arch and full-arch custom
tray impressions. This time, however, the authors used a
heavy body impression material in the tray, and no mention was given to the order in which the dual-arch impressions were poured. Mean marginal openings were
between 25 and 28 m in all test groups, and they
concluded that there was no statistical difference in the
marginal fit of castings made with full-arch custom
acrylic trays, metal, or plastic dual-arch trays.
Breeding and Dixons9 in vitro study investigated the
dimensional accuracy of casts as a function of the type of
dual-arch tray used and the type of impression material
used. It was concluded that metal trays produced smaller
144

Fig. 1. Stainless steel standard on typodont. Mesiodistal, buccolingual, and occlusogingival dimensions were measured
for standard and resultant working dies. Dimensions shown
for standard.

dies than the typodont tooth and plastic trays produced


larger dies. However, the authors poured the working
side of the dual-arch impressions first and did not pour
the opposing arch at all.
Distortion is a 3-dimensional problem, and when it is
applied to a study on impression material accuracy, several factors can be analyzed.10-12 With dual-arch trays
there is a lack of consensus about which impression tray
should be used, what viscosity of impression tray material is needed, and which side of the dual-arch impression should be poured first to minimize distortion. Investigation of these parameters would also help in
planning an anticipated clinical trial.
The purpose of this investigation was to compare the
accuracy of working dies made from impressions with
metal and plastic dual-arch trays when 2 different viscosities of impression tray material were used, and altering
which side of the dual-arch impression was poured first.
The null hypotheses was that there was no difference in
die materials for plastic versus metal trays, for
monophase versus rigid impression material, and for
pouring opposing versus the working side of the impression first.

MATERIAL AND METHODS


Impressions were made of the right side of a typodont
containing a machined, circular stainless steel crown
preparation (standard) in the position of a mandibular
right premolar (Fig. 1). Three variables that could affect
accuracy of working dies were studied: type of dual-arch
tray used, viscosity of the impression material for the
tray, and the order of pouring the 2 arches of the impression. A balanced design with independent samples
was used to study these 3 variables. A sample size of 10
was used, yielding a total of 80 impressions.
VOLUME 90 NUMBER 2

CEYHAN, JOHNSON, AND LEPE

THE JOURNAL OF PROSTHETIC DENTISTRY

Fig. 2. A Plastic and B metal dual-arch impression trays used in investigation.

Two common types of dual-arch trays were used: a


plastic tray (Triple Tray; Premier Dental Products Co,
Plymouth Meeting, Pa.) and a metal tray (COE Impression Tray; GC America Inc, Alsip, Ill). Two consistencies of polyvinylsiloxane impression material,
monophase and rigid, were used in the impression trays
(Aquasil Rigid and Aquasil Monophase; Dentsply
Caulk, Milford, Del). Aquasil XLV was always used for
injection onto the standard stainless steel die before the
tray was seated (Fig. 2). The working (preparation) side
was poured first for half the impressions, and the opposing side was poured first for the remaining impressions.
The dual-arch trays were assessed to ensure that the
typodont could be closed into maximum intercuspal position without any interference from the tray. A dual-mix
technique was used where tray and low-viscosity materials were automixed simultaneously. With a disposable
syringe (Ultradent Products, South Jordan, Utah), 1.2
mL of light-bodied material was injected around and
over the stainless steel standard, and 5 complete activations of the auto-mix cartridge was used to dispense 11
mL of the tray material (Aquasil Rigid or Aquasil
Monophase; Dentsply Caulk) into both sides of the dual-arch trays. The dual-arch trays were seated onto the
mandibular right quadrant, the typodont was closed and
a 1.5-kg weight placed on top to simulate a constant
occlusal force.6,7 A jig was used so that the position of
the impression tray was constant and reproducible between trials (Fig. 3). The impressions were removed 12
minutes after the start of mix, twice the manufacturers
setting time, to compensate for the temperature of the
extraoral environment.11 The impressions were rinsed
under tap water for 10 seconds, dried, and poured in
gypsum 60 minutes later. Type IV gypsum (Fuji-Rock;
GC America), with a ratio of 20 mL of distilled water to
100 g of powder, was hand-mixed for 10 seconds and
then mixed under vacuum for 40 seconds and poured
into the trays while being vibrated. One side of the dualarch impression was poured with 35 g of stone and
allowed to set for 1 hour before the other side was
poured with 35 g of stone. The tray handles were placed
AUGUST 2003

Fig. 3. Impression tray held in fixed position with typodont in


closed position with 1.5-kg weight in place.

in a holder to ensure the impression was suspended for


the 60 minutes allocated for the gypsum to set. The
order of pour was randomized, and all casts were allowed to set for 24 hours at room temperature before
removal.
The casts were then sectioned with a diamond disc
(6934; Brasseler USA, Savannah, Ga) to form individual
gypsum dies of the stainless steel standard. The gypsum
working dies were placed into a custom jig that was
fabricated to permit their measurement at a fixed, reproducible position under the microscope (Fig. 4). Three
aspects of each die were measured (buccolingual, mesiodistal, and occlusogingival) with a measuring microscope (Nikon Measurescope; Nikon, Tokyo, Japan) so
that a 3-dimensional representation of accuracy could be
assessed. Each dimension of the working dies was measured 3 times, and the mean was used for the sample
value. The same 3 aspects of the stainless steel standard
were measured multiple times, first before and then at
the conclusion of measuring all working dies, to arrive at
the 3 standard values to which all working die means
were compared. The mean values for the standard used
in the statistical analysis were those taken at the conclu145

THE JOURNAL OF PROSTHETIC DENTISTRY

Fig. 4. Stainless steel jig used to ensure standard and gypsum


dies are measured in constant, reproducible position.

Fig. 6. Factor level mean values for difference of mesiodistal


dimension of working die and standard, for comparison of
trays and sequence of pouring.

CEYHAN, JOHNSON, AND LEPE

Fig. 5. Factor level mean values for difference of buccolingual dimension of working die and stainless steel standard
(mm).

Fig. 7. Mean difference of mesiodistal dimension of working


die and stainless steel standard (mm) for monophase and
rigid, for both trays.

cosity of the impression material for the tray


(monophase, rigid), and pouring sequence of the casts
(opposing, working). The sample size was 10 and all
hypothesis testing was conducted at .05. Because
there were 2 groups within each factor, there was no
need to conduct post hoc tests for mean differences since
any significant main effect (P.05) would indicate a
statistically significant difference between the 2 groups.

RESULTS
Fig. 8. Factor level mean values for difference of occlusogingival dimension of working die and stainless steel standard
(mm).

sion of the study. The intraexaminer variation for measuring the standard was 0.001 mm.
A 3-factor analysis of variance was used to analyze the
data. The factors were type of tray (metal, plastic), vis146

The mean buccolingual, mesiodistal, and occlusogingival dimensions of the stainless steel standard and the
working dies, for all 3 factors, are given in Table I. In
general, the gypsum dies were smaller than the stainless
steel standard in all 3 dimensions. The mean standard
deviation associated with the stainless steel standard was
0.001 mm and that for the working dies ranged from
0.005 to 0.013 mm. To better illustrate how the dimensions of the working dies compared with the standard,
for the 3 factors of study, the net difference of the 3
VOLUME 90 NUMBER 2

CEYHAN, JOHNSON, AND LEPE

THE JOURNAL OF PROSTHETIC DENTISTRY

Table I. Mean values and standard deviations for all combinations


MONO
STD

Plastic
Buccolingual
Metal
Plastic
Mesiodistal
Metal
Plastic
Occlusogingival
Metal

RIGID

Oppose

Work

Oppose

Work

7.053 (.005)

7.053 (.013)

7.050 (.007)

7.051 (.006)

7.051 (.005)
7.041 (.008)

7.054 (.007)
7.043 (.012)

7.049 (.005)
7.048 (.009)

7.048 (.005)
7.051 (.007)

7.064 (.017)
6.243 (.007)

7.066 (.014)
6.243 (.008)

7.046 (.011)
6.251 (.005)

7.053 (.010)
6.248 (.007)

6.244 (.013)

6.247 (.009)

6.250 (.008)

6.255 (.007)

7.071 (.001)

7.071 (.001)

6.250 (.001)

Table II. Results of 3-factor ANOVA for buccolingual dimensions (BL)


Tests of between-subjects effects

Source

Type III sum


of squares

df

Mean square

Sig.

Noncent.
parameter

Observed
powera

Corrected Model
Intercept
TRAY
ARCH
VISC
TRAY*ARCH
TRAY*VISC
ARCH*VISC
TRAY*ARCH*VISC
Error
Total
Corrected Total

7.203E-03b
1989.680
3.497E-03
3.844E-04
7.921E-04
2.250E-05
2.132E-03
3.721E-04
3.600E-06
2.623E-03
1989.689
9.826E-03

7
1
1
1
1
1
1
1
1
32
40
39

1.029E-03
1989.680
3.497E-03
3.844E-04
7.921E-04
2.250E-05
2.132E-03
3.721E-04
3.600E-06
8.197E-05

12.553
2.4E07
42.658
4.689
9.663
.274
26.003
4.539
.044

.000
.000
.000
.038
.004
.604
.000
.041
.835

87.871
24271784
42.658
4.689
9.663
.274
26.003
4.539
.044

1.000
1.000
1.000
.556
.854
.080
.999
.542
.055

Dependent variable: BL.


a
Computed with .05.
b 2
R .733 (Adjusted R2 .675).

dimensions of the working die and the standard are plotted in Figures 5 through 8. All cross-product interactions were not significant (P.05), except the viscosity/
tray interaction for the mesiodistal dimension (P.05)
(Tables II, III, and IV). Thus, factor-level or grand
means are plotted where interactions were not significant (Figs. 5, 6, and 8), and individual mean values are
shown for the situation in which there was a significant
interaction (Fig. 7).
There was a statistically significant mean difference
for type of impression material viscosity used in the tray.
The use of a rigid impression material resulted in gypsum dies that were significantly smaller than the standard in a buccolingual dimension (0.021 mm, P.047)
and taller in an occlusogingival dimension (0.001 mm,
P.004) (Figs. 5 and 8).
In the mesiodistal dimension, there was a significant
interaction of tray type with impression material viscosity
(P.05). When a monophase impression material was
used, plastic dual-arch trays yielded gypsum dies that were
AUGUST 2003

significantly smaller (0.029 mm) than the ones generated


from the metal trays (0.006 mm) (Fig. 7).

DISCUSSION
Distortion is a 3-dimensional problem that is inherent in all of the steps involved in fabricating an indirect
dental restoration.10 The intra-abutment distortion of a
crown preparation, and the distortion of the die in a
mesiodistal, buccolingual, and occlusogingival direction
was investigated in this study. The change in the interabutment distances that occur among multiple crown
preparations and the surrounding teeth was not addressed in this study. The clinical relevance of this study
is that the distortion observed could be extrapolated to
that of a single crown preparation, but not to one involving multiple preparations.
The gypsum dies obtained from the impressions in
this study were generally smaller in all 3 dimensions than
the standard. This may be due to the effect of not having
147

THE JOURNAL OF PROSTHETIC DENTISTRY

CEYHAN, JOHNSON, AND LEPE

Table III. Results of 3-factor ANOVA for mesiodistal dimensions (MD)


Tests of between-subjects effects

Source

Type III sum


of squares

df

Mean square

Corrected
Model
Intercept
TRAY
ARCH
VISC
TRAY*ARCH
TRAY*VISC
ARCH*VISC
TRAY*ARCH*VISC
Error
Total
Corrected Total

.458b
1965.464
6.037E-02
6.416E-02
7.413E-02
6.131E-02
6.480E-02
6.257E-02
7.073E-02
.767
1966.689
1.225

7
1
1
1
1
1
1
1
1
32
40
39

6.544E-02
1965.464
6.037E-02
6.416E-02
7.413E-02
6.131E-02
6.480E-02
6.257E-02
7.073E-02
2.398E-02

2.729
81975.575
2.518
2.676
3.092
2.557
2.703
2.610
2.950

Sig.

.024
.000
.122
.112
.088
.120
.110
.116
.096

Noncent.
parameter

19.105
81975.575
2.518
2.676
3.092
2.557
2.703
2.610
2.950

Observed
powera

.834
1.000
.337
.355
.400
.342
.358
.347
.385

Dependent variable: MD.


a
Computed with .05.
b 2
R .374 (Adjusted R2 .237).

Table IV. Results of 3-factor ANOVA for occlusogingival dimensions (OG)


Tests of between-subjects effects

Source

Type III sum


of squares

df

Mean square

Corrected Model
Intercept
TRAY
ARCH
VISC
TRAY*ARCH
TRAY*VISC
ARCH*VISC
TRAY*ARCH*VISC
Error
Total
Corrected Total

5.762E-04b
1557.816
1.225E-06
1.102E-05
2.652E-04
2.403E-05
2.256E-04
3.802E-05
1.103E-05
2.271E-03
1557.819
2.847E-03

7
1
1
1
1
1
1
1
1
32
40
39

8.231E-05
1557.816
1.225E-06
1.102E-05
2.652E-04
2.403E-05
2.256E-04
3.802E-05
1.103E-05
7.098E-05

1.160
2.2E07
.017
.155
3.737
.338
3.179
.536
.155

Sig.

Noncent.
parameter

Observed
powera

.353
.000
.896
.696
.062
.565
.084
.470
.696

8.118
21948799
.017
.155
3.737
.338
3.179
.536
.155

.416
1.000
.052
.067
.46
.087
.409
.109
.067

Dependent variable: OG.


a
Computed with .05.
b 2
R .202 (Adjusted R2 .028).

used a tray adhesive in this study. During the polymerization reaction, the impression material shrinks toward
the center of mass. The use of a tray adhesive, however,
would redirect this shrinkage toward the impression tray
walls, resulting in a die larger in diameter but shorter in
height.11 In the absence of a tray adhesive there would
be unrestricted polymerization shrinkage of the impression material, resulting in a die that is smaller in diameter
and height. Tray adhesive was not used in this study
because both tray types had mechanically retentive features. Introducing a tray adhesive would have incorporated another variable because the taller sides of the
metal tray would have received more adhesive than the
shorter sides of the plastic tray. Breeding and Dixon9
148

found similar results with the metal dual-arch trays,


where dies were undersized but could not explain why
the plastic dual-arch tray impressions yielded dies that
were oversized. However, they only studied the buccolingual dimension of the dies and only poured the working side of the impression. It is possible the weight of the
gypsum in the plastic trays distorted the unsupported
impression because the opposing impression was never
poured. In this investigation it was noted that when the
monophase tray material was used with the plastic dualarch trays, there was a larger standard deviation for all 3
dimensions when the working side of the impression was
poured before the opposing. This may explain the protocol of Wilson and Werrin,1 of always pouring the
VOLUME 90 NUMBER 2

CEYHAN, JOHNSON, AND LEPE

counter-impression before pouring the working side impression.


The machined stainless steel standard used in this
investigation provided certain advantages in obtaining
the measurements over that of a prepared plastic typodont tooth. The well-defined line angles of the stainless
steel standard were clearly observed under the microscope, thereby reducing measurement error. This could
explain the smaller standard deviations observed for
measuring the standard in this investigation (1 m)
compared with the study of Breeding and Dixon where
the standard was a prepared typodont tooth (16 m).9
The Davis and Schwartz investigations6-8 used a casting
cemented on a prepared typodont tooth, and the results
were only presented as percentage change from the standard. The standard deviations observed were larger than
the means in most instances. The circular nature of the
standard allowed observation of the relationship between the change in buccolingual and mesiodistal dimensions of the gypsum dies. With the plastic trays, the
gypsum dies tended to change from a circular shape into
an ovoid shape, they were generally narrower mesiodistally than buccolingually.
When extrapolating the data to a clinical situation, it
should be kept in mind that the results are applicable
only to a single tooth situation and only to the types of
impression trays and materials tested. Different trays and
impression materials may yield different results. The effects of the lips, tongue, and cheeks in containing the
impression material could not be simulated in this study,
nor could the influence of an occlusal force in excess of
the 1.5 kg used in this investigation (Fig. 3). Finally, this
investigation yielded some statistically significant differences between tray type and impression material viscosity, however, the differences were of a magnitude that
would probably have little clinical significance. The
thickness of 1 coat of die spacer has been shown to vary
from 8 to 40 m.12 Consider the mesiodistal group
mean measurements, when monophase tray material
was used, between plastic and metal trays (Table I). The
greatest difference in the group mean values occurred
here (23 m); however, this difference would be insignificant clinically with the application of die spacer.

CONCLUSION
Within the limitations of this study the following conclusions were made:
1. The gypsum dies produced from the dual-arch
impressions were generally smaller in all 3 dimensions
than the stainless steel standard.

AUGUST 2003

THE JOURNAL OF PROSTHETIC DENTISTRY

2. The buccolingual dimension of the gypsum dies


made from the rigid impression material was slightly
larger than the standard.
3. The mesiodistal dimension of the gypsum dies
generated from the plastic dual-arch trays was smaller
than the ones generated from the metal trays when a
monophase impression material was used.
4. The occlusogingival dimension of the gypsum dies
generated from the rigid impression material was larger
than the monophase generated gypsum dies.
5. Although statistically significant differences were
found, the magnitude of these differences is clinically
insignificant since the difference can be compensated for
with 2 coats of die relief.
The authors thank Dentsply Caulk for supplying the impression
material and Premier Dental Products Company for donating the
plastic dual-arch trays.

REFERENCES
1. Wilson EG, Werrin SR. Double arch impressions for simplified restorative
dentistry. J Prosthet Dent 1983;49:198-202.
2. Bass EV, Kafalias MC. Dual-arch impressions. Aust Dent J 1992;37:1-5.
3. Kaplowitz GJ. Trouble-shooting dual arch impressions. J Am Dent Assoc
1996;127:234-40.
4. Kaplowitz GJ. Trouble-shooting dual arch impressions II. J Am Dent Assoc
1997;128:1277-81.
5. Imbery TA. Simplifying dual-arch impressions. J Am Dent Assoc 1998;
129:1599-600.
6. Davis R, Schwartz R, Hilton T. Marginal adaptation of castings made with
dual-arch and custom trays. Am J Dent 1992;5:253-4.
7. Davis RD, Schwartz RS. Dual-arch and custom tray impression accuracy.
Am J Dent 1991;4:89-92.
8. Schwartz RS, Davis RD. Accuracy of second pour casts using dual-arch
impressions. Am J Dent 1992;5:192-4.
9. Breeding LC, Dixon DL. Accuracy of casts generated from dual-arch
impressions. J Prosthet Dent 2000;84:403-7.
10. Nicholls JI. The measurement of distortion: theoretical considerations. J
Prosthet Dent 1977;37:578-86.
11. Johnson GH, Craig RG. Accuracy of four types of rubber impression
materials compared with time of pour and a repeat pour of models. J
Prosthet Dent 1985;53:484-90.
12. Campbell SD. Comparison of conventional paint-on die spacers and those
used with the all-ceramic restorations. J Prosthet Dent 1990;63:151-5.
Reprint requests to:
DR GLEN H. JOHNSON
DEPARTMENT OF RESTORATIVE DENTISTRY
BOX 357456
UNIVERSITY OF WASHINGTON
SEATTLE, WA 98195-7456
FAX. 206-543-7783
E-MAIL: gjohnson@u.washington.edu
Copyright 2003 by The Editorial Council of The Journal of Prosthetic
Dentistry.
0022-3913/2003/$30.00 0

doi:10.1016/s0022-3913(03)00276-2

149