Content uploaded by Alvin Wee
Author content
All content in this area was uploaded by Alvin Wee on Feb 15, 2018
Content may be subject to copyright.
BASIC SCIENCE RESEARCH
Evaluation
of
the Accuracy
of
Solid
Implant Casts
Alvin
G.
Wee,
BDS, MS,'
Robert
L.
Schneider,
DDS,
,US,2
Steven
A.
Aquilino, DDS,
MS,'
Thomas
L.
HuJ
DDS,
MS,4
Tey
J.
Lindquist,
DDS,
MS:
and
Derrick
L.
Williamson, DDS,
MS5
Purpose:
Materials used to fabricate the most dimensionally accurate implant casts have not
been identified experimentally. The purpose of this investigation was to examine the dimensional
accuracy of implant casts fabricated with different materials. Measurements
of
linear horizontal
dimensional change and strain produced on a master framework were evaluated and correlated.
Materials and Methods:
A master framework was fabricated to
fit
an aluminum five-implant
model. Forty polyether implant impressions of the aluminum model were randomly grouped and
poured
in
either Vel-mix, Die Keen, Resin Rock, or Low Fusing Alloy. A digital veneer caliper was
used to measure linear distance between the most distal abutments on each of the experimental
implant casts and the master model. In addition, strain values were recorded from strain gauges
bonded in the mesiodistal axis
of
the framework, which was secured by prosthetic retaining screws
torqued to
10
Ncm.
Results:
A one-way ANOVA showed a significant difference among the four die materials
in
dimensional change
of
the experimental casts
(p
=
.0001).
A post-hoc Duncan's multiple-range test
(p
<
.05)
showed that casts fabricated
with
Low Fusing Alloy had the least linear dimensional
change from the master cast, but the material exhibited the greatest dimensional variability. A
MANOVA (Wilks' Lambda) showed significant differences
in
strain on the framework based upon
die material
(p
=
.015).
A post-hoe Duncan's multiple-range test
(p
<
.05)
showed that Resin Rock
casts induced significantly less strain on the framework than the other materials. Negligible
correlation was found between the linear horizontal dimensional change and the total absolute
strain on the framework.
Conclusion:
Experimental implant casts made
of
Resin Rock minimized strain on the master
framework and decreased the amount of framework distortion on casts of this material. Low
Fusing Alloy yielded accurate casts, but highly variable linear dimensional changes in the horizontal
dimension may preclude its clinical benefit.
J
Prosthod 1998;7:161-169. Copyrighto 7998
by
The American College
of
Prosthodontists.
INDEX
WORDS:
dental implants, dental materials, dental gypsum, dental casts, dental alloys
'Assidant
Projissor,
Section ofRestoratioe Dentisty, ProJthodontirs and
2Associate
Pmfwor,
Department OJProstthodontics and Clinical Director,
Endodontics,
The
Ohio State Unicu~@, Columbur, OH.
Oral and Maxillofaacial Jmplant Center, The Universi@ ofIowa, College
of
Dentistry, Iowa
Ci@,
01.
3Pro~sor and Director
of
Graduate Prosthodontics, Department
of
Prosthodontics,
The
LTnicm'@ OJIowa,
College
offintisty, Iowa
CiQ,
,!A.
4Asociate
Projssor
and
Head,
Defiartment
ofPrmthodontics,
Chiuersity
OJTexm
at
Houston,
Houston,
TX.
5A.rsirtant
Projissor,
Department
of
Prosthodonhs,
The
C'niuusity
of
Iowa,
College
ofDentistgi, Iowa
Cip,
M.
Accepted
May
28,
1998.
Sufiwed
in part
b
a
1996
Greatu New
Ymk
Academji ofhsthodon-
tics Stua'ent Grant and donation ofthe implant components
t=g~
Nobel
Biocure
(USA).
This
articl,?
UI~J
one
ofthzjnalistsjfir
thz
199RArthurR. Frechette
New
Inuestigafm Award and
war
presented
in
part at the
1998
American
AssociationJor Dental Rmarch Annual
Session.
Correspondence to:
Dr.
Alvin G. M-e, Section
of
Restoratioe Dentist?).:
Prosthodontics and Endodontics,
College
dDentGtv,
I'oste
Hall,
305
West
12th
Avenue,
Columbur,, 01-143210-1241.
Copyright
0
1998
ig
i%e
American
ColleKe
ofPrmthdontists
1059-941X/98/
O7O3-0003$500/ 0
LTHOUGH
osseointegrated titanium implants
A
used in dentistry today enjoy a high success
rate,'" clinical complications exist that may lead to
delayed implant component failure. Reported compo-
nent failures include loosening
of
prosthetic retain-
ing screws, fracturing or locking
of
abutment retain-
irig screws, and fracturing of the prosthesis or
implant^.',*,^-^
Konpassive fit
of
the implant-retained
prosthesis on the abutments may be a reason for
delaycd component failures.4,GJ0-'2
Distortion during master
cast
fabrication, one
of
the many phases
of
prosthesis fabri~ation,'~ could
contribute to a misfit
of
the definitive prosthesis to
the implant abutments when implant-retained fixed-
detachable hybrid prostheses
are
fabricated indi-
rectly. The implant cast must reproduce the adjacent
hard and soft tissues and accurately represent the
intraoral positional relationship
of
the implant abut-
ments. The accuracy
of
the cast
is
dependent on the
impression and the implant master
Journal
OfProsthodontics,
Vol7,
No
3
(Sebtenibed
1998:pf
161-169
161
162
Amray ofsolid Implant
Casts
a
Wee
et
a1
cast techniq~ie.l~~Z~~2~;'~ However, the effect
of
differ-
ent die materials on implant cast accuracy has not
been vigorously investigated.
Expansion of the gypsum products used to fabri-
cate the implant cast could alter the positional
relationship ofthe implant abutment replicas. Ameri-
can Dental Association
(ADA)
Type
IV
or
V
dental
stones are recommended,13 but numerous materials
with varying linear expansions could be used to
fabricate the solid implant cast. Ideally, the expan-
sion should compensate for polymerization changes
of the impression material. Nevertheless, a compari-
son of die materials with different linear expansion
on the accuracy of an implant master cast fabricated
using polyether impression material is unknown.
A
multitude
of
methods has been used
to
measure
distortion
of
master casts and
fit
of implant prosthc-
ses. For instance, a common method to assess accu-
racy of the cast fabrication process has bccn to
measure either the three-15J3 or
two-dimensior~al~~~~~~~~
linear translation
of
the implant components with a
traveling microscope. Others have used three-
dimensional distortion analysis to evaluate the move-
ment
of
the gold cylinders under
a
variety of condi-
tion~.~~,~~.~~-~~ Still others have used gold cylinder-
abutment replica gap distancc measureinents,IH,'"~,~
Yeriotest measurements,j* and torque turn analy-
sis."5 Measurement of the linear distance between
the framework and implant cast
16~1i,23,3G,37
and deter-
mination of thc amount of stress on either the
implant framework'0.24 or a simulated resin man-
~lible~~J~ have
also
been used.
Although many measures of misfit have been
reported, it
is
uncertain how these measuremcnts
are associated and/or correlated.
If
the thrcc-
dimensional distortion in positioning the gold cylin-
ders during framework fabrication causes misfit
be-
tween gold cylinders and the iritraoral abutments,
internal stresses
will
occur within the prosthesis-
implant-bone complex.'
'
Measures of linear horizon-
tal dimensional change between the most distal
abutment replicas
of
a solid implant cast and mea-
sures of strain on the secured implant framework on
that cast provide valuable information regarding
three-dimensional distortion. These indices of
fit
can
be readily obtained, quantified, and analyzed within
a
clinical or research setting, and any relationship
between such measurcs could provide additional
insight regarding the nature arid manifestation of
stress within the prosthesis-implant-bone complex.
The purpose of this study
was
threefold:
1)
to
evaluate the linear horizontal dimensional change
of
experimental implant casts fabricated with dlfferent
materials compared with
an
implant master model;
2)
to assess the mesiodistal strain produced on
a
master framework secured to the experimental im-
plant casts
by
prosthetic retaining screws torqued to
10 Ncm; and
3)
to
dctermine if a correlation exists
between the horizontal linear dimensional change
and the total absolute strain on the secured frame-
work.
Materials
and
Methods
This in vitro study investigated the accuracy of materials
uscd to fabricatc solid implant casts.
A
master framework
was fabricated
to
fit an implant master model. Tmplant
impressions were made of thc master modcl and poured in
one of four die materials (Table
1).
-4
digital veneer cdiprr
was used to measure thc linear horizontal dimensional
change of each experimental cast compared with the
master model. Strain values werc
also
recorded on the
framework after it was secured to the various experimental
casts
by
prosthetic retaining screws.
An implant master model was milled out of a solid
aluminum block mith five stainless-steel abutment replicas
(DC-4
174,
Nobelpharma USA Inc, Chicago,
IL)
prcss-
fittcd into the model (Fig
1).
The master framework was
waxed and cast in a high-pdlladium alloy (Porcelain
76,
W.E.
Mowrey Company, St. Paul, MNj.
To
cnsure a passive
fit, the abutmcnt replicas were removed from the master
model, secured to the framework by prosthetic retaining
screws (DCA
075,
Nobelpharma,
LX4
Inc),
aid
torqued to
10 Ncm with
a
hand torquc driver (DIA
250,
Blue -10
Ncm, Nobelpharma
US.4,
Inc). Holes in the aluminum
model were enlarged until the abutment replicas attached
to the framework could be positioned passively into the
preparcd holes (Fig 2).
An
adhesive resin cement (T'anavia
21,
Dental Adhesives, Kureray Dental
J,
Morita, Tustin,
CA) was uscd according to the manufacturer's instructions
to
cement the abutment replicas into the prepared holes.
The complcx was lcft undisturbed for
1
week. Mter visual
confirmation of framework fit, stainless-steel sct screws
Table
1.
Experiment Materials anti Batch Piumbers
~
Products Batch Xumbers
Die Keen
Impregum Adhesivr
Impregum Penta ESPE
Low
Fusing
Alloy
Microstone
Prosthetic Retaining Screws
liesin
Kock
Strain Gauges
(E1A-06-0624P-
Triad
Vel-mix
1204ption
P)
9508326
049/0047
2862
1
no batch number
072796008
3461
1
05549
6002
S
14.5
153
960828B
W%
I
002
I
275
September
1998,
Volume
7,
Number
3
163
Figure
1.
Aluminum implant master model milled out
of
an aluminum block. The abutment replicas
are
labeledA4
to
E
from
Ltj?
to
righl.
were drilled and fastened horizontally through the abut-
ment replicas
to
secure them in place.
Forty polyether impressions (Impregum
F,
Premier
Dental Products Go, Norristown, PA) were made of the
aluminum master model using individualized trays made
of light-polymerized resin (Triad, Dentsply Int Inc, York,
PA). Ten impression tray3 and
10
sets
of
five direct square
impression copings (DCB
026,
Nobelpharma USA Inc)
were distributed randomly in the experiment. Impression
copings were secured to the master model by guidc pins
torqued to
10
Ncm. The polyether impression material
(Impregum
F,
Premier) was mixed in an electric mixing
and dispensing machine (Pentamix,
ESPE,
Premier), and
an impression
was
made
of
the implant master model (Fig
3).
The manufacturer’s suggested setting time of
6
min-
utes was doubled to compensate for polymerization at
room temperature rather than mouth temperature.“”
,4
digital timer was used to standardize each step of the
procedure.
Imprcssions were poured in one of four dic materials.
These materials were as follows: 1) Vel-mix, an ADA type
IV
dental stone (Kerr Manufacturing Company, Emery-
Figure
2.
“Framcworlg‘abutment rrplica” complex be-
fore cementation with Panavia
2
1
to
the
master model.
Figure
3.
Implant master model impression.
ville,
CA)I3;
2)
Die Kccn, an ADA type V dental stone
(Miles Dental Products, South Bend,
IN)’3;
3)
Resin
Rock,
a
resin-impregnated gypsum hybrid matcrial (Whip Mix
Corporation, Louisville,
KY)’O;
and
4)
Low Fusing Alloy
(Belmont Alloy
2491,
Silver, Belmont Metals Inc, Brooklyn,
NY)
with Microstone, an ADA type
ILI
dental stone (Whip
Mix
Corporation)?l
All
gypsum products were preweighed
using a digital scale. They were thcn stored in heat-sealed
plastic bags until
use.
The
products were vacuum-mixed
with distilled water according to the manufacturers’ recom-
mendations.
Stainless-steel abutment replicas
(DC4
174,
Nobel-
pharma USA Inc) were hand-tightened with 15-mm guide
pins to the five square impression copings in each impres-
sion. Impressions were poured
30
minutes after removal
from the master model to simulate the clinical situation.
Experimental casts were allowed to set for
1
hour before
unscrewing the <guide pins and removing the impression.
One experimental cast was poured from each impression,
and
a
total
of
40
experimental implant
casts
were fabri-
cated.
To fabricate the cxperimcntal
casts
made of
Low
Fusing Alloy (melting temperature:
136”F),
the material
was preheated in
a
150°F
water bath in
a
12-cc plastic
syringc until it was in
a
liquid statc. The syringe was
used
to
dispense
5
cc of Low Fusing Alloy into the impression to
cover the abutment replicas. The
1,ow
Fusing Alloy was
allowed
to
ct~l for
5
minutes
before
placing the retentive
components (Fig
4).
Thc remaining portion of the imprcs-
sion was poured in Microstone.
‘l’he
40
experimental casts were storcd in ambient
conditions for at least
24
hours42 before linear or strain
measurements were made. ‘I’he
10
experimental casts
within a material group were numbered and randomly
distributed
to
form
10
sets
of
four casts such that each set
contained one cast made from each material. The linear
horizontal dimensional measurements and strain measure-
ments were performed according to the sequence illus-
trated in Fig
5.
Each measurement on cach experimental
cast and the master model was made on three separate
occasions.
164
Accuracy
ofSolid
Implant
Cmts
Wee
et
a1
Figure
4.
Implant impression intaglio surface after being
poured with
Low
Fusing Alloy.
During the pilot study, the precision of the digital
veneer caliper
was
found to be 0.02 mm, and the precision
of the strain measurements was found to be 20
pe.
Subsequently, if any of the three repeated readings for the
linear dimensional change or strain measurements were
not within the calculated precision range, the readings
were made again until three recordings for that particular
cast or strain gauge fell within the range.
A digital veneer caliper (Digimatic Caliper, model
CD-6 BS, Mitutuyo Corporation,
Tokyo,
Japan) was posi-
tioned in
a
repeatable fashion
on
each experimental cast
and the master model with a four-point contact
on
the two
most distal abutment replicas. The distance between the
most distal abutment replicas was measured and recorded
(Fig
6).
Four general-purpose, unidirectional strain gauges with
preattached lead wires (EA-06-06”-120 with Option P,
Micro-measurements Division, Measurement Group Inc,
Raleigh,
NC)
were bonded to the occlusal aspect of the
master framework with the strain gauge axis oriented
mesiodistally. One strain gauge was bonded between each
of the five gold cylinders (DCA
075,
Nobelpharma
USA
/
START
4
,
Measurement instrument
was
zeroed.
+
+
Readings were taken
on
the master cast.
i
-
C
(10times1
411
ten
%eo
of four implant
cast5 Here measured
(\=a).
Whole
sequence was repeated
two additional times on separate
occasions, resulting in three
repeated measures for
each
implant cast.
4
END
Measurement instrument was zeroed.
Readings were taken on one cast
from the four included in
a
set.
Each cast in theset
was
fabricated
from one ofthe four different
tested materials.
1
One set offour experimental
casts
was
measured.
Figure
5.
Measurement sequence for experimental im-
plant casts.
Inc) (Fig
7).
The
strain gauge jumper wircs were connected
to the appropriately numbered, color-coded binding post
of
a Companion Switch and Balance Unit (Model SB-10,
Instruments Division, Measurement Group Inc). The bal-
ancing unit was then connected to the Model
P-3500
Strain
Indicator (Instruments Division, Measurement Group Inc)
to create a quarter bridge, 120-0hm configuration (Fig
8).
Each of the four indibidual strain gauges was balanced to
50000
strain with
no
load
on
the framework before
measurements were taken.
The
framework was secured to
the various experimental casts using prosthetic retaining
screws torqued to 10 Ncm with a hand torque driver in a
standardized sequence. When it was activated, the strain
indicator displayed the reading for the appropriate strain
gauge. A dfferent set of prosthetic retaining screws was
used for each
of
the three repeated strain measurements.
The absolute strain value used in the analysis for an
example strain gauge
(SG
01)
was calculated using the
formula (A
-
B
=
[C]), where
A
was the mean of the three
repeated strain values
of
SG
01
on an experimental
implant cast, B was the mean of
30
repeated strain values
of SG 01
on
the aluminum implant master model, and
C
was the actual strain value. For
A
and
B,
positive and
negative values were used to calculate the means. Only the
positive values
of
C
were used for statistical analysis. Thus,
C
was the absolute strainvalue.
Statistical analyses were performed using the
SAS
statistical program?3
A
one-way
ANOVA
(a
=
0.05)
was
used to evaluate the statistical significance oflinear horizon-
tal dimensional change among the four material groups.
Thereafter, a post-hoc Duncan’s multiple-range test was
used to rank the means of the experimental groups. A
MANOVA
(a
=
0.05)
using the Wilks’ Lambda test was
used to evaluate the statistical significance of the mean
absolute strain values of the four strain gauges among and
ihithin the four material groups. The two factors for the
strain data were:
1)
Material, the “between-subjects” fixed
effect; and
2)
Position, the “within-subject” or repeated
fixed effect. Thereafter, a post-hoc Duncan’s multiple-
range test was used to compare mean strain values among
die material groups at each strain gauge position. Finally,
for each material,
a
Pearson’s correlation coefficient
(r)
was
used
to
determine the relationship between the mean
horizontal linear dimensional change for each cast and the
total absolute strain recorded by the four strain gauges on
the framework.
Results
Linear Horizontal Measurements
The results
of
the linear horizontal measurements
are reported in Table
2.
Values represent the mean
difference
of
the distance between the terminal
Sebtember
1998,
Volume
7,
Number3
165
Figure
6.
Digital veneer
cali-
per
used
to
measure
the
most
distal abutments
of
the casts.
abutments on the implant master model
(43.86
+
0.007
mm) and each of the experimental groups. The
one-way
ANOVA
(p
=
.0001)
revealed a statistically
significant difference among the experimental casts
based on die material type (Table
3),
and the Dun-
can’s multiple-range test (Table
4)
showed
no
diffcr-
ence in linear measurement betwccn Resin Rock and
Vel-mix
groups. Nevertheless, Die Keen casts had
significantly greater horizontal dimensional change
than the other groups, and
Low
Fusing Alloy
had
significantly less dimensional change. The mean
horizontal linear measurement from
Low
Fusing
Alloy experimental casts did not diffcr from thc
master model, but the coefficient
of
variance was
20
to
70
times greater than the other groups (Table
2).
Strain
Gauge
Values
on
the
Framaomk
A
significant effect was found based
on
an interaction
between the die material type and the strain gauge
position (Table
5).
Table
6
illustrates the mean
absolute strain for each strain gauge among the four
die material groups. For three
of
the lour strain
gauge positions, casts madc of Resin Rock induced
the least amount of strain on the framework. In
addition,
no
statisticdly significant difference in
mean absolute strain was found among groups from
SG
03
and
SG
04.
Nevertheless, statistically signifi-
cant differences in mean absolute strain were identi-
fied, and implant casts
of
Resin Rock induced signifi-
cantly less strain on the framework compared
with
Figure
7.
Four
strain gauges
bonded
to thc master frame-
work.
166
Accuray
of
Solid
Implant
Casts
Wee
et
al
casts of Vel-mix at
SG
01 and
Low
Fusing Alloy at
SG
02.
Correlation
of
the
Linear Measurements
and
Strain Vdues
Pearsonian correlation coefficients
(r)
were calcu-
lated, and negligible correlation was found between
the total absolute strain on the framework and the
mean horizontal linear dimensional change. The
r2
values comparing linear measurements and strain
values for the different experimental groups were as
follows: Vel-mix
=
0.09, Die Keen
=
-0.12,
Resin
Rock
=
-0.01, and
Low
Fusing
Alloy
=
0.14.
There-
fore, linear dimensional change was not an accurate
predictor
of
strain within the framework.
Discussion
The results
of
this study agree with Tan et
aI,@
who
stated that the magnitude
of
linear distortion mea-
sured in most studies is not directly related to the
actual stress induced within the implant system
concerned. Negligible correlation between linear hori-
zontal dimensional change measurements at the
Figure
8.
Strain measure-
ment system with strain indi-
cator, balancing unit, and
strain gauge on the master
framework.
terminal abutment and total absolute strain measure-
ments on the framework shows that linear dimen-
sional change does not adequately predict the strain
placed on the secured implant framework in the
mesiodistal direction. Therefore, in distortion analy-
sis, it
is
important to target measurements of three-
dimensional distortion or a clinically relevant mani-
fcstation of distortion, such
as
stress within the
prosthesis-implant-bone complex, to derive a rel-
evant conclusion.
The standard deliation of the mean for the three
repeated measurements was acceptable for both the
linear horizontal dimensional change (0.01 mm) and
strain measurements
(8
~LE)
The standard deviation
of repcated measurements for individual strain
gauges in this study was similar to those reported
by
Assif et aP4
(7
p),
in which similar methodology
was
used. Although every attempt was made
to
standard-
ize thc multiple variablcs in thc expcriment, possible
explanations for the variability among the material
groups (Table
6)
include:
1)
the four materials have
their
own
range of expansion;
2)
polyether may
interact differently with each of the four materials
used;
3)
the expansion in the three dimensions may
differ for each
of
the four materials;
4)
the machining
Table
2.
Linear Horizontal Dimensional
Charige
of
Experimental Casts
Mean
SD
Range
%
hfference
CoeJ%ent
Malerial
N
(mm)
(mm)
SEA4
immi
From
iMaster
Cast
of
Variance
Vel-mix
10 0.02 0.01 0.05 0.00-0.04 0.04 88
Die Keen
10 0.04 0.01 0.04 0.02-0.06 0.10
28
Kesin
Kock
10 0.02 0.02 0.05 -0.01-0.04 0.04 I00
Low
Fusing
Alloy
10 0.00 0.02 0.06 -0.02-0.04 0.00 1960
K'ote.
Means
were
calculated
using
the
mean
of
thc
three
rcpeated
rcadmgs
from
each
cast
September
1998,
Volume
7,
iVumber
3
167
Table
3.
One-way ANOVA
@
<
.05)
for linear
Horizontal Dimensional Change Measurements
Table
5.
MANOVA
ci,
<
.05)
for Mean Absolute Strain
on the Secured Mastcr Framework
Sum
of
df
Squares
F
Value
P
~~~~~
Model
3
0.01004
12.60
.om1
Error
36
0.00956
tolerances of the implant components varied4’; and
5)
the use of only three
sets
of prosthetic retaining
screws may have resulted in deformation
of
the
screws during the study.
The variability of Low Fusing Alloy dimensional
measurements could also be a result of its partial
eutectic microstructure and multiple alloy composi-
tion. Flow or creep may have occurred at room
temperature, because the recrystallization tempera-
ture
of
the matrix metal, lead, is very
In
this
experimental design, three repeated measurements
werc
made at differcnt times, and flow or creep could
have occurred bctwecn measurements. Nevertheless,
Low
Fusing
Alloy
had significantly greater mean
absolute strain values only at the
SG
02
position.
If
greater strain occurred at other positions, it may not
have been detected because the gauges measured
strain in only one dimension.
Although the framework fit to the aluminum
implant master model was visually acceptable, the
four strain gauges detected strain on thc framework
when the implant framework was secured to the
model. This was taken into consideration for data
analysis through the use of absolute strain values.
The sign in front of the strainvalue is an indication of
the “direction”
of
strain, not the “amount” of strain.
In strain gauge technology, thc “negative” sign indi-
cates compression, and the “positive” sign indicates
tension.
A
negative strain value does not imply less
strain than a positive strain value. If the absolute
values were used to calculate
A
(the mean
of
the
strain measurements on the experimental casts) and
B
(the mean of the strain measurements on the
Table
4.
Duncan’s Multiple-Rangc Tcst
ci,
<
.05)
for
Linear Horizontal Dimensional Change
Mean Duncan?
Material
N
fmml
Grou$ing*
Die Keen
10 0.04
mm
A
Vel-mix
10 0.02
mm
B
Resin Rock
10 0.02
mm
B
hw
Fusing
Alloy
10
0.00
mm
c
*Identical
letters
in
the same column reprerent
means
that
are
not
significantly dillerent.
Value
F
Value
df
p
Wilks’ Lambda
0.561 2.47
9
.015
82.9
aluminum model), then
C
(the actual strain) would
not be a true representation of actual strain. The
mean actual strain would therefore not represent the
spectrum of both compression and tension readings,
and inappropriately identified significant differences
could be found during statistical analysis. Therefore,
the absolute value of
C
was appropriate to use for
subsequent analysis.
The positions
of
the strain gauges were arbitrarily
determined on the occlusal surface of the implant
framework. Studies that evaluate implant cast accu-
ra~y’1~;~’ have shown that the horizontal dimensional
change is more significant than vertical change.
Therefore, the strain gauges were placed in the
rnesiodistal direction, and implant cast horizontal
dimcnsional change measurements were made. In
the future, photoclastic stress analysis during
a
pilot
study could bc used to locate areas on the framework
where a hlgh degree
of
strain occurs. Strain gauges
could then be selectively bonded in the mesiodistal
dircction in these areas
on
the implant framework.
A
more relevant position to place strain gauges is
around the implant abutments to form a load cell, as
reported byJemt.q7 The level of strain at the implant
abutment would be clinically important, because
most implant complications occur within the pros-
the tic retaining scre\v/implant abutment/abutment
screw comple~.~~~~~-~ When components are stressed
beyond their long-term fatigue capacity, delayed
component failure results. The implant abutment
could be calibrated, and with the known modulus of
elasticity, stress induced in the abutment replica
could be calculated. If the resultant distortion were
measured using a three-dimensional analysis? bend-
ing moments could also be calculated. This could not
be accomplished with the present experimental
model.
An alternative method to measure stress on the
implant framework would be to use photoelastic
stress analysis as reported by Uludamar and Leung.48
A
photoelastic plastic coating could be bonded onto
the superior surface
of
the implant framework, and
a
reflection polariscope would be used to view the
fringe order on the secured framework. Greater
168
Accuracy
of
Solid
Implant
Casts
Wee
et
a1
Table
6.
Duncan’s Multiple-Range Test for Mean Absolute Strain
@)
for Each Strain Gauge
on
the Secured
Master Framework
Strain Gauge
01
Strain Gauge
02
Strain Gauge
0.7
Strain Gauge
04
Material Mean
SD
SEM
Croups*
itlean
SD
SEM
Groups*
Mean
SD
SEM
Groups* Mean
SD
SEM
Groups*
Vel-mix
47.30 54.25 17.16
A
35.20 26.22 8.29
AB
25.00 27.97 8.84
A
21.20 24.18 7.55
4
DieKeen
25.40 15.02
4.75
AB
26.00 17.19 5.44
B
37.10 27.47 8.69
A
25.80 18.20
5.76
A
Low
Fusing
Alloy.
17.80 12.63 3.99
B
44.60 13.73 4.97
A
16.60 17.86 5.65
A
17.00 13.82 4.37
A
ResinRock
18.50 9.76 3.08
I3
22.60 16.95 5.36
€3
16.00 12.33 3.90
A
11.50 11.55 3.65
A
*Duncan’s groupings following
MANOVA
(Multivariate
Wilks’
Lambda) using ranked data. Identical letters in the same column represent
means that are not significantlydiffererit
fringe ordcr is correlated with greater stress on the
framework.
Clinical Significance
Compared to Vel-mix, Die Keen or Resin Rock,
Low
Fusing Alloy used
in
this experiment produced the
least change in linear dimension for a five-implant
model when it was poured in
a
polyether impression.
However, because
of
the exceptionally high variance
of Low Fusing Alloy measurements and its more
complex technique, it may not
be
clinically predict-
able. Resin Rock produced the least mean absolute
strain on the framework in three of
the
four strain
gauges. Therefore, frameworks fabricated on
a
solid
implant cast made of Resin Rock may produce less
strain on the implant framework in the mesiodistal
direction compared with casts made of Vel-mix, Die
Keen, and
Low
Fusing Alloy. Use of Resin Rock may
reduce the amount of stress in the prosthesis-implant-
bone complex. The upper limit of the mean strain
value range for Resin Rock
was
found to be
22.6
p~.
Using
Hook’s
law, if the high-palladium implant
framework’s Young’s modulus were estimated at
95
bPa
X
lo3, the stress gcneratcd would be
2.15
ma.
Although this amount of stress may
be
minimal, its
clinical relevancc has yet to be determined.
Conclusion
The accuracy of solid implant casts can be influenced
by the type of materials used to fabricate them.
Statistically significant differences both in linear
horizontal dimensional change and strain on the
framework were found among the four materials
tested. Resin Rock produced the second-most dimen-
sionally accurate solid implant casts and the least
amount of strain on the implant framework in this
study. Therefore, Resin Rock may be
a
practical
alternative
to
fabricate
solid
implant casts.
Acknowledgment
The authors thank
Dr.
Clark Standford for his
review
of
the
protocol,
Dr. ’lhomas
Southard
for
his
insight
on
the
use
of
strain gauges, and Dr. Supanee Buranadham
for
her
constant support and input during the
course
of
this
project.
’They
also
thank
Mrs.
.Jane
Jacobsen and Dr.
William Johnston (The Ohio
State
University) for their
statistical support during the analysis of the data.
References
Albrektsson
T
A
multicenter report on osseointegrated oral
implants.
J
Prosthet Dent
1988;60:75-84
,4dell R, Eriksson
B,
Lekholm
C,
ct a1
A
long term follow up
study
of
osseointegrated implants in the treatment of totally
edentulous
jaws.
Int
.J
Oral hlaxillofac Implants
1990;5:347-
359
Zarb GA, Schmitt A The longitudinal clinical effectiveness of
osseointegrated dental implants. The Toronto study. Part
11:
The prosthetic results.
J
Prosthet Dent
1990a;6453-51
AdeU
R,
Lekholm
U?
Rockler
B,
et al:
A 15-year
study of
osseointegrated implants in the treatment of the edentulous
jaw. IntJOralSurg
1981;10:387-416
5.
Johansson G, Palmcpist
S:
Complications, supplementary
treatment and maintenance in edentulous arches with im-
plant supported futed prostheses. Int
J
Prosthodont
1990;3:
6.
Jemt
T
Failures and complications
in
391
consecutively
inserted fmed prostheses supported byBr%nemai-k implants in
edentulous
jaws:
A
study of treatment from the time of
prosthesis placement to the first annual checkup. Int
J
Oral
Maxillofac Implants
1991;6:270-276
7.
Naert
I,
Quiqnm
hf,
Van Steenberghe
D,
et
al:
A
study of
589
consecutive implants supporting complete fixed prostheses.
Part
I1
Prosthetic aspects..J Prosthet Dent
1992;68:949-956
8.
Tolman L)E, Laney
WIZ:
Tissue-integrated prosthesis compli-
cations. Int
J
Oral
Maxillofac Implants
1992;7:477-484
9.
Carlson
B;
Carlsson GE: Prosthodontic complications in ossee
integrated dental implant treatment. Int
J
Oral Maxillofac
Implants
1994;990-94
10.
CollinsJW,
Rigsby DF, Dixon DL, et
al:
Accuracy of working
casts Tor
an
implant supported restoration
(abstr).
J
Dent Res
19Y6;7j:
I6
1
1
I.
Skalak
R
Biomechanical considerations in osseointegrated
prostheses.
J
Prosthet Dent
1983;49:843-848
89-92
September
1998,
Volume
7,
Number
3
169
12.
Kallus
T,
Bessing C: Loose gold screws frequently occur in
full-arch fEed prostheses supported by osseointegrated im-
plantsafter
5years.IntJOralh.ZaxillofacImplants
1994;9169-
I78
13. Zarb GA, Jansson
TP:
Prosthodontic procedures and labora-
tory procedures and protocol, in Brhemark PI,
Zarb
G&
Alberktsson I, (eds): Tissue Integrated Prostheses. Chicago,
IL:
Quintessence, 1985, pp 241-282,293-315
14. Humphries RM, Yaman P, Bloem TJ: The accuracy of implant
master casts constructed from transfer impressions. Int
J
Oral
Maxillo Implants 1990;5:331-336
15. Spector
M,
Donovan
TE,
NichollsJI:
An
evaluation of imptes-
sion techniques for osseointegrated implants.
J
Prosthet Dent
1990;63:444-447
16.
Carr
AB:
A comparison of impression techniques for
a
five-implant mandibular model. Int
J
Oral
Maxillofac lm-
plants 1991;6:448-455
17.
Carr
AB:
Comparison of impression techniques for
a
two-
implant 15 degree divergent model. Int
J
Oral Maxillofac
Implants 1992;7:468-475
18.
AssXD, Fenton
AH,
Zarb GA, et
al:
Comparative accuracyof
implant impression procedures. Int
.J
Periodont Kest Dent
1992;lZ:
1
13-12 1
19.
Barrett
MG,
de Rijk WG, BurgessJO The accuracy of
six
impression techniques for osseointegrated implants.
J
Prosthod
1993;2:75-82
20. Inturregui JA, Aquilino
SA,
Ryther JS, et al: Evaluation
of
three impression techniques for osseointegrated oral
im-
plants.
J
Prosthet Dent 1993;69:503-509
21.
Liou
AD,
NichollsJ, Yuodelis
RA,
et
al:
Accuracy of replacing
three tapered transfer impression copings into
two
elasto-
meric impression materials. Int
J
Prosthodont 1993;6:377-383
22.
Phillips
KM,
NichollsJI, Ma T, et
al:
The accuracy
nf
three
implant impression techniques: A three-dimensioinal analysis.
Int
J
Oral Maxillofac Implants 1994;9:533-540
23. Carr
AB,
MasterJ: The accuracy of implant verification casts
compared with casts produced from
a
rigid transfer coping
technique.
J
Prosthod 1996;5:248-252
24. Assif' D, Marshak B, Schmidt
A
Accuracy of implant impres-
sion techniques. Int
J
Oral Maxillofac Implants 1996;11:216-
222
25.
Hsu
CC, Millstein PL, Stein
RS:
A
comparative analyTis
of
the
accuracy of implant transfer techniques.
J
Prosthet Dent
26.
Hussaini S, Wong
T
One dinical visit for
a
multiple implant
restoration master cast fabrication.
J
Prosthet Dent
1997;78:
550-553
27.
Vigolo P, Millstein
PL:
Evaluation of master cast techniques
for multiple abntment implant prostheses. Int
J
Oral hhillo-
fac Implants 1993;8:439-445
28.
Alwazzan
K,
Ziebert
G,
Balthazar
Y
The accuracy of
a
removable die system for implant frameworks (abstr).
J
Dent
Res 1994;73:232
29. Rodney
J,
Johansen R>
Harris
W:
Dimensional accuracy of
two
implant impression copings (abstr).
J
Dent Res 1991;
70(special
issue):385
1993;69:588-593
30.
Ness
EM,
Nicholls
JI, Rubenstein
JE,
et
al:
Accuracy of the
acrylic rcain pattern for the implant retained prosthesis. Int
J
Prosthcdont 1992;5:542-549
31. Tan
KB,
RubensteinJE, NichollsJE, et al: Three dimensional
analysis
of
the casting accuracy of one-piece, osseointegrated
implant retained prostheses. Int
J
Prosthodont 1993;6:346-363
32. ,Jemt T: In
vivo
measurements of precision of
fit
involving
33.
34.
35.
36.
37.
38.
39.
49.
41.
42.
43.
44.
45.
46.
47.
on
osseointegrated implants supporting fured or removable
prostheses: A comparative pilot study. lnt
J
Oral blaxillofac
Implants 1991;6:413417
48. Uludamar
.4,
Leung
T
Inaccurate
fit
nf
implant superstruc-
tures. Part
IL
Efticacy
of
the precidisc system for the correc-
tion
or
errors. IntJProsthodont 1996;9:16-20
implant-supported prostheses in the edentulous jaw. Int
J
Oral
Maxillfac Implants 1996;11:151-158
Cheshire PD, Hobkirk.JA An in vivo quantitative analysis
of
the lit
of'
Nobel Biocare implant superstructures.
J
Oral
Kehabil 1996;23:782-789
May
KB,
Edge
MJ,
Lang BR, et
al:
The periotest method:
Implant-supported framework precision of fit evaluation.
J
Prosthod 1996;5:206-2 13
Wicks
RA,
de Rijk WG, Windeler
AS:
An evaluation of fit in
osseointegrated implant components using torquehrn analy-
sis.
J
Prosthod 1994;3:206-212
Carr
AB,
Gerard DA, Larsen PE: The response of bone in
primates
arourd
unloaded dental implants supporting prosthe-
ses with different levels offit.
J
Prosthet Dent 1996;76:500-509
Carr
AB,
Stewart
RB:
Full-arch implant framework casting
accuracy. Preliminary in vitro observation for in
viva
testing.
J
Prosthod 1993;2:2-8
Clelland
NL,
Can
AB,
Gilat
A
Comparison of strains trans-
ferred to
a
bone simulant between =-cast and postsoldered
implant frameworks
for
a five-implant-supported fixed prosthe-
sis. JProsthod 1996;5:193-200
Clelland
NL,
Van Putten MC: Comparison of strains pro-
duced in
a
bone simulant between conventional cast and
resin-luted implant frameworks. Int
J
Oral Maxillofac Im-
plants 1997;12:793-799
American Dental Association: Revised American Dental
Assc-
ciation specification no. 19 for non-aqueous, elastomeric den-
tal
impression materiltls. JAm Dent Assoc 1977;94:733-741
Schneider
RL,
Wee AG: Fabricating of low fusing metal casts
for more accurate implant prosthodontics.
J
Prosthod 1996;5:
301-303
Arlo
K
Evaluation of the Vident die and model system.
USAF
Dental Investigation Services, 1991
SAS
Institute
I:
SAS
User's Guide: Basic (5th ed). Cray,
NC,
SAS Institute Inc, 1985
Tan
KB:
The clinical significance of distortion in implant
prosthodontics:
Is
there such
a
thing
as
passive fit? Ann Acad
Med 1995;24: 138-157
Binon
PI?
The effect of implant/abutment hexagonal misfit
on screwjoint stability. Int
J
Prosthodont 1996;9:149-160
Anusavice
JK:
Phillip's Science of Dental Materials (10th ed).
Philadelphia, PA, Saunders, 1996
Jemt T, Chrlsson
L,
Boss A, et
al:
In vivo load measurements
