CLASSIC ARTICLE

The effect of various finish line preparations on the marginal seal and
occlusal seat of full crown preparations
J. R. Gavelis, DMD,a J. D. Morency, DMD,b E. D. Riley, DMD,c and
R. B. Sozio, DMDd
University of Connecticut, School of Dental Medicine, Farmington, Conn; Naval Air
Station Hospital, Corpus Christi, Tex; and Harvard School of Dental Medicine, Boston, Mass

T he full veneer crown is one of the most important

restorations in the armamentarium of the restorative
then poured improved dental stone directly into the
castings to form the dies.
dentist. The restoration can restore severely deteriorated The purpose of this study was to correlate margin
teeth. One problem recognized by clinicians is that the design with the seating and sealing of cemented full
cementing medium may prevent the seating of the full cast crowns under standardized, simulated clinical
crown, positioning it in hyperocclusion and causing conditions.
inadequately sealed margins. Investigators report the
film thickness of cement along the axial walls of the METHODS AND MATERIAL
preparation influences the seating of the restoration.
The factors of cementation pressure,1-3 duration of Eight stainless steel dies were machined to produce
cementation,2,3 powder/liquid ratio of the cement,2,4-6 a crown preparation similar in volumetric size to an
preparation dimensions,1,2,4 type of cement,6,7 occlusal average molar. Seven dies were 10 mm in diameter and
perforations,1,2 die spacers,7 and relief of the internal the eighth was 8 mm in diameter at the finishing lines.
crown surface4,8 have been related to the film thickness. Each die measured 6 mm from the occlusal surface to the
One study related the type of finish line of the end of the preparation, and each had a 5-degree taper of
preparation to the film thickness. Fusayama et al3 the axial walls with a convergence angle of 10 degrees.
cemented cast full crowns onto extracted teeth prepared Two dies had featheredge finish lines, one 8 mm in
with the 90-degree shoulder, the 45-degree shoulder, diameter and the other 10 mm. The other margin
and featheredge margins. They reported that the designs were: a 1 mm 90-degree shoulder; a 45-degree
featheredge margin provided the best sealing effect shoulder; a 90-degree shoulder and a chamfer with
followed by the 45-degree shoulder and 90-degree bevels 1 mm long parallel to the axial walls; a 1 mm 90-
shoulder, respectively. In an earlier study, Fusayama degree shoulder with a 45-degree bevel; and a 1 mm
et al9 found that, even without cement, crowns fail to 90-degree shoulder with a 30-degree bevel (Fig. 1).
seat completely owing to the numerous variables Five crowns were fabricated from each die. The
associated with the casting process. crowns were waxed on the steel dies, invested (Beauty
Teteruck and Mumford10 also reported similiar Cast; Whip Mix Corp, Louisville, Ky), and cast
findings while comparing castings made with various (Firmilay; J. F. Jelenko, New Rochelle, NY) in type III
gold alloys and investments. All investigations in which gold (Fig. 2). Because of the casting variables reported in
patterns were waxed and cast and the castings returned previous studies,9,10 a method similar to that described
to the original dies were affected by the casting variables. by McCune6 was used to fabricate dies directly into the
In an effort to eliminate the casting error, McCune6 castings. After the castings were pickled and the sprues
fabricated castings on the Bureau of Standards’ die and removed, the castings were boxed with masking tape.
The castings were lubricated with a fluorocarbon dry
release agent (Crown Industrial Products Co, Hebron,
Ill) and a measured volume of methyl methacrylate
Presented as a table clinic before the American Academy of Crown (Duralay; Reliance Dental Mfg. Co, Chicago, Ill) was
and Bridge Prosthodontics, Chicago, Ill.
a
Assistant Professor, Department of Restorative Dentistry, University vibrated into the castings to make direct dies. Twenty-
of Connecticut, School of Dental Medicine. four hours later, the dies were undercut below the finish
b
Staff Dentist, Naval Air Station Hospital. line to facilitate measurements and indexed to aid proper
c
Associate Professor, Department of Prosthetic Dentistry, Harvard repositioning at cementation (Fig. 3).
School of Dental Medicine.
d The crowns were cemented onto the Duralay dies
Associate Professor and Head, Department of Graduate Prosthetic
Dentistry, Harvard School of Dental Medicine.
with an Instron testing machine (Instron Corp, Canton,
Reprinted with permission from J Prosthet Dent 1981;45:138-45. Mass) (Fig. 4, A). An orange wood stick was placed
J Prosthet Dent 2004;92:1-7. between the compression head of the testing machine

JULY 2004 THE JOURNAL OF PROSTHETIC DENTISTRY 1

THE JOURNAL OF PROSTHETIC DENTISTRY GAVELIS ET AL

Fig. 1. Diagram of standard die used in study, with various finish line preparations in inserts.

Fig. 2. Crowns were waxed on steel dies (A) and cast in a type III gold alloy (B).

and the crown to simulate clinical practice and to dis- the crown with a diamond disk on an ultrastructural
tribute the cementation pressure more evenly (Fig. 4, B). analysis cutting machine (Buehler Ltd, Evanston, Ill)
Initially cementation pressure was 100 pounds, but this (Fig. 5). The cut surfaces were wet polished with
was slowly reduced to 57 pounds over the 10-minute progressively finer grit sandpaper, 240 to 600 grit,
cementation interval. Then the cemented crowns were toward the crown to eliminate flash. The cement spaces
embedded in plastic and sectioned through the center of were measured to the nearest micron with an eyepiece

2 VOLUME 92 NUMBER 1
GAVELIS ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

Fig. 3. Dies were indexed to facilitate repositioning during cementation (A) with arrows and (B) with a depression and
corresponding dimple.

Fig. 4. A, Crowns were positioned for cementation with an Instron testing machine. B, An orange wood stick was used to
distribute pressure evenly during cementation.

JULY 2004 3
THE JOURNAL OF PROSTHETIC DENTISTRY GAVELIS ET AL

Fig. 5. Cemented crowns were sectioned with diamond disk (A) through center of crown (B).

Fig. 7. Arrows show orientation and location where cement

lines were measured.

half of the sectioned crowns were averaged to arrive at

thickness values of the cement line at the margin,
shoulder, axial wall, and occlusal surface. The cement
thicknesses at the margin and occlusal surface were
analyzed to find the amount of seal and seat afforded by
the various preparations. Statistical analysis consisted of
one-way analysis of variance with multiple comparisons
using Tuckey’s test and Scheffe’s test. Significance was
determined at the P\.05 level.
Fig. 6. Photomicrograph of cement spaces at margins.
The data in Table I and Fig. 8 show that the
micrometer mounted on a reflecting microscope featheredge preparations had the best marginal seal, 31
(Fig. 6). and 34 m. The parallel bevel preparations followed, with
marginal seals of 41 and 44 m. No statistical difference
was found between the featheredge and parallel bevel
RESULTS
preparations. The 90-degree shoulder had a cement
Measurements were made of the cement lines at space of 67 m at the margins. The 45-degree shoulder,
points shown in Fig. 7. The measurements from each shoulder with 30-degree bevel, and shoulder with

4 VOLUME 92 NUMBER 1
GAVELIS ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

Fig. 8. Comparison of cement space at margins and occlusal surface of various finish line preparations.

Table I. Marginal seal and occlusal seating as measured in microns

Marginal seal Occlusal seat

Marginal design Measurement (m) SEM SD Measurement (m) SEM SD

Featheredge, 10 mm diameter 31 6 3.30 7.4 163 6 12.66 28.3

Featheredge, 8 mm diameter 34 0.98 2.2 157 8.81 19.7
45-degree shoulder 95 7.65 17.1 138 12.66 28.3
90-degree shoulder 67 10.20 22.8 85 14.49 32.4
Shoulder, parallel bevel 41 0.72 1.6 214 12.48 27.9
Chamfer, parallel bevel 44 3.44 7.7 196 15.07 33.7
Shoulder, 30-degree bevel 99 11.72 26.2 170 18.29 40.9
Shoulder, 45-degree bevel 105 15.83 35.4 153 25.13 56.2

45-degree bevel followed, with spaces of 95, 99, and shoulders with 45-degree and 30-degree bevels. The
105 m, respectively. There was no significant difference differences between members of these groups, the 90-
between these three finish lines, but the differences degree shoulder group, and the group with parallel
between this group and the 90-degree shoulder group bevel preparations were significant.
and the featheredge and parallel bevels group were
significant.
DISCUSSION
The measurements at the occlusal surface found the
shoulder preparation restorations to be the most By using standardized steel dies of volumetric size
completely seated with an 85 m cement space. Ideally, similar to that of teeth, rather than prepared natural
the space at the margin and that at the occlusal part of teeth, castings of duplicate size and shape are produced,
the shoulder preparation should be identical. The thus decreasing the variation within each group of finish
discrepancy in these areas was found not to be line preparations. The machined dies also enable the
statistically significant. The order of seating after the investigator to accurately control the variables of prep-
shoulder preparation was the 45-degree shoulder at 138 aration dimensions, degree of axial wall taper, and the
m, the shoulder with 45-degree bevel at 153 m, the finish line dimensions. The Bureau of Standards’ die was
featheredges at 157 and 163 m, the shoulder with 30- not utilized because it is 15 mm in diameter, which is
degree bevel at 170 m, the chamfer with parallel bevel at much larger than a natural molar, and the convergence
196 m, and the shoulder with parallel bevel at 214 m. angle of the Bureau of Standards is 3 degrees, which is
Differences were not significant between the follow- too severe for the clinical situation. Textbooks recom-
ing groups: 45-degree shoulder, featheredges, and mend a convergence angle of 4 to 14 degrees.11,12 For

JULY 2004 5
THE JOURNAL OF PROSTHETIC DENTISTRY GAVELIS ET AL

geometry of the preparations. Most of the crowns

seated obliquely on the dies, a finding also reported by
Jorgensen1 and Bassett.8 As geometrically predicted, the
featheredge margin had the best marginal seal. There
was no significant difference between the featheredge
preparations of 8 mm and 10 mm diameters. Ranked
next were the parallel bevel margins with acceptable13
marginal seals in the 40 m range.
A finding which was not consistent with other
studies3,11,12,14 was that the 90-degree shoulder prep-
aration ranked after the parallel bevel margins with an
opening of 67 m. The 45-degree shoulder and shoulders
with 30-degree and 45-degree bevels were the worst
seals, with openings ranging from 95 to 105 m. For
geometric design, the shoulder would be expected to
have the greatest opening at the margins. This finding is
explained by evaluating the measurements of the
occlusal space.
The 90-degree shoulder preparation had the best
seat with an 85 m cement space. The other prep-
arations followed, with the parallel bevel preparations
the worst observed with 196 to 214 m cement spaces.
The seating of the 90-degree shoulder preparation
improved the marginal seal. However, the featheredge
and parallel bevel preparations were superior in sealing
the margins despite the poor seating. The differences
in measurements of occlusal seating is an interesting
finding since the only variable was the type of finish
line. If the degree of seating were solely dependent on
Fig. 9. Control showing no shrinkage of acrylic resin away the film thickness of the axial wall cement, all
from crown. specimens would have seated to the same degree
occlusally. This indicates that the variation of the
these reasons, the dies were machined to a convergence finish line is related to the different degrees of crown
angle of 10 degrees with a 5-degree taper on each axial placement.
wall. The question is how the finish lines affect cementa-
Perhaps the most significant source of variation tion. When the crown is cemented the axial wall of the
between the results of previous studies3-6,8-10 stems preparation approaches the axial wall of the internal
from casting variables. This source of error was crown surface. The escape path for the cement
controlled with acrylic resin dies poured directly into decreases, causing the hydrostatic pressure within the
boxed castings. It is accepted that methyl methacrylate crown to increase until it matches the patient’s biting
resin shrinks on polymerization. However, this factor pressure. At this point, the crown fails to seat further. If
did not affect the present study. Controls were the cement does not set, it will continue to escape until
established by pouring dies into castings, followed by the particles at the axial walls prevent further seating.
embedding (unseparated) and sectioning 24 hours after Certain finish lines apparently facilitated the escape of
polymerization. Microscopic examination revealed no cement early in the cementation process under the
shrinkage spaces between the die and the internal surface conditions of the study.
of the crown (Fig. 9). Shrinkage did occur, but it was It is thought that less hydrostatic pressure results in
limited to the open end of the boxed castings, thus greater seating of the restorations.
causing meniscus formation. However, the explanation is not simple. The filtration
To further relate this study to the clinical situation, process observed by Jorgensen1,2 and Petersen15 is
the crowns were cemented with an orange wood stick thought to contribute to the differences observed in this
placed between the crown and the compression head of study. Hoard et al16 demonstrated that peak hydrostatic
the Instron testing machine. The stick aided in pressures are short-lived and are redistributed into
dispersing the cementation forces more evenly. intracoronal pressures. They concluded that this leads
The findings of this study supported those of to various localized filtration processes which affect the
previous studies and the predictions based on the flow of cement and final seating of the crown.

6 VOLUME 92 NUMBER 1
GAVELIS ET AL THE JOURNAL OF PROSTHETIC DENTISTRY

The better seating of the castings with shoulder REFERENCES

preparations is due to the poor seal prior to complete 1. Jorgensen KD. Factors affecting the film thickness of zinc phosphate
cements. Acta Odontol Scand 1960;18:189.
cementation. This design allows the cement to escape
2. Jorgensen KD. Structure of the film of zinc phosphate cements. Acta
marginally more readily without filtration. With the Odontol Scand 1960;18:491.
other finish line preparations, the margins seal earlier 3. Fusayama T, Ide K, Hosoda H. Relief of resistance of cement of full cast
crowns. J Prosthet Dent 1964;14:95.
and start filtration sooner. With the parallel bevel
4. Kaufman EG, Coelho DH, Colin L. Factors influencing the retention of
preparations, the comparatively large number of internal cemented gold castings. J Prosthet Dent 1961;11:487.
angles produces the greatest amount of filtration and 5. Kaufman EG, Colin L, Schlagel E, Coelho DH. Factors influencing the
poor seating. Venting,1,2 die spacing,7 and internal relief retention of cemented gold castings: The cementing medium. J Prosthet
Dent 1966;16:731.
of the crown4,8 help to reduce filtration and enable the 6. McCune, R. J. The effect of occlusal venting and film thickness on the
parallel bevel preparations to seat more completely. cementation of full cast crowns. [thesis] Indianapolis (IN): Indiana
University, School of Dentistry; 1968.
In relating this study with actual clinical practice it
7. Eames WB, O’Neal SJ, Monteiro J, Miller C, Roan JD Jr, Cohen KS.
must be remembered that the dies fit the crowns per- Techniques to improve the seating of castings. J Am Dent Assoc 1978;
fectly. In clinical practice the casting variables may have 96:432.
8. Bassett RW. Solving the problems of cementing the full veneer cast gold
greater effect on the fit of the crowns than the variation
crown. J Prosthet Dent 1966;16:740.
of the finish lines. Further studies are needed to evaluate 9. Fusayama T, Ide K, Kurosu A, Hosoda H. Cement thickness between cast
the filtration process and hydrostatic pressures within restorations and preparation walls. J Prosthet Dent 1963;13:354.
10. Teteruck WR, Mumford G. The fit of certain dental casting alloys using
the crown. Studies also should be done to correlate the
different investing materials and techniques. J Prosthet Dent 1966;
increases in hydrostatic pressure in crown restorations 16:910.
with different finish lines with the seating of the crowns. 11. Tylman SD. Theory and practice of crown and fixed partial prosthodon-
tics. 6th ed. Philadelphia: Mosby; 1971. p. 223.
12. Johnston JF, Phillips RW, Dykema RW. Modern Practice in Crown and
Bridge Prosthodontics. 3rd ed. Philadelphia: W.B. Saunders; 1971. p. 68.
SUMMARY 13. Christensen GJ. Marginal fit of gold inlay castings. J Prosthet Dent
1966;16:297.
The influence of the marginal design of a full crown 14. Pascoe DF. Analysis of the geometry of finishing lines for full crown
restorations. J Prosthet Dent 1978;40:157.
on the occlusal seat and marginal seal of a cemented full 15. Jorgensen KD, Petersen GF. The grain size of zinc phosphate cements.
crown restoration was examined. Under the conditions Acta Odontol Scand 1963;21:255.
of the study, the featheredge and parallel bevel prep- 16. Hoard RJ, Caputo AA, Contino RM, Koenig ME. Intracoronal pressure
during crown cementation. J Prosthet Dent 1978;40:520.
arations demonstrated the best marginal seal, followed
in order by the full shoulder, 45-degree shoulder, and
finally the 90-degree shoulders with 30-degree and 45- 0022-3913/$30.00
degree bevels. With regard to seating of the restoration, Copyright ª 2004 by The Editorial Council of The Journal of Prosthetic
the 90-degree full shoulder demonstrated the best seat, Dentistry

followed in order by the 45-degree shoulder, 90-degree

shoulder with 45-degree bevel, featheredge, 90-degree
shoulder with 30-degree bevel, chamfer with parallel
bevel, and finally 90-degree shoulder with parallel bevel. doi:10.1016/j.prosdent.2004.03.024

JULY 2004 7